Container crane control system and method based on situation awareness

Abstract

The application discloses a container crane control system and method based on situation awareness, relates to the crane control technical field, collects standard data of the crane, establishes a historical database about the characteristics of the crane after preprocessing and normalization, searches the historical database, selects the crane for executing the task by screening the same kind of goods tasks, analyzes the historical data, splits the task links, analyzes the task link completion degree, judges the link execution state, analyzes the characteristic deviation condition, thereby analyzes the influence degree of the characteristic on the link according to the characteristic deviation condition, comprehensively considers the task completion degree, thereby obtains the correlation coefficient of the link execution state on the task completion condition, analyzes the adaptation degree of the crane to the task, obtains the preselected crane, comprehensively considers the residual working time length of the preselected crane, selects the crane to receive the hoisting task, and the application controls the real condition of the container crane through the situation awareness, formulates the preventive measures, and enhances the overall operation stability.

Description

Technical Field

[0001] This invention relates to the field of crane control technology, specifically to a situational awareness-based container crane control system and method. Background Technology

[0002] Currently, most methods in container crane control rely on preset parameters, failing to dynamically adjust based on the actual operating status of the equipment and the characteristics of the cargo. These methods, based on fixed specifications, ignore performance deviations caused by wear and tear, environmental changes, and other factors during long-term use. This makes it difficult to accurately match the actual capabilities of the equipment with the task requirements when handling different types of cargo, leading to fluctuations in operational efficiency and even increasing the risk of task execution anomalies. This fails to meet the precision and flexibility requirements of modern industry. During container crane operations, task execution anomalies frequently occur, but current technology lacks systematic correlation analysis of the root causes. When deviations occur or tasks fail, it is difficult to determine whether the cause is an abnormality in equipment characteristics or a problem in the operation itself. Faults can only be diagnosed based on single phenomena, failing to establish an effective correlation between equipment characteristics, operational steps, and task completion. This results in low efficiency in anomaly troubleshooting, an inability to develop targeted preventative measures, and the recurrence of similar anomalies, affecting overall operational stability. Furthermore, current container crane task allocation primarily considers the equipment's idle state, without scientifically assessing the compatibility between the crane and the task. The allocation process ignores key factors such as the historical performance and stability of equipment, which may result in tasks being assigned to idle but poorly adapted cranes. This not only reduces the efficiency of single task execution but may also increase the probability of failure due to equipment and task mismatch, affecting the smoothness of the overall port operation process. For example, the invention patent with patent number CN118917527A discloses "A method for energy consumption management and prediction of container port cranes and its medium", which describes a management method for container port cranes. However, it only uses fixed parameters and considers the task allocation of idle cranes. There are problems such as lag in crane control parameters, difficulty in tracing anomalies, and coarse allocation. It is difficult to provide a guarantee for the efficient and safe operation of port cranes and is not suitable for refined operation scenarios under complex working conditions. Summary of the Invention

[0003] The purpose of this invention is to provide a situational awareness-based container crane control system and method to solve the problems mentioned in the background art.

[0004] To address the aforementioned technical problems, this invention provides the following technical solution: a container crane control method based on situational awareness, comprising the following steps:

[0005] S1. Collect standard data of cranes, and after preprocessing and normalization, establish a historical database of crane characteristics;

[0006] S2. Search the historical database, filter similar cargo tasks to obtain the cranes that perform the tasks, analyze the historical data, break down the task steps, and analyze the completion rate of the task steps;

[0007] S3. Determine the execution status of the process, analyze the characteristic deviation, and then analyze the degree of influence of the characteristic on the process based on the characteristic deviation.

[0008] S4. Analyze the impact of the execution status of each step on the task completion status, and comprehensively consider the degree of task completion to obtain the correlation coefficient between the execution status of each step and the task completion status.

[0009] S5. After receiving the task in real time, retrieve historical data, analyze the adaptability of the crane to the task, and obtain the pre-selected crane.

[0010] S6. Taking into account the remaining working time of the pre-selected cranes, select a crane to accept the lifting task.

[0011] Further, in step S1, standard data of the crane is collected. This standard data is the data specified in the crane's manufacturer's specifications. Based on situational awareness, real data of the crane is collected. This real data is the data obtained based on situational awareness during crane operation. After preprocessing and normalization, standard characteristics of the crane are obtained through the standard data. For example, the standard data includes the response time of any type of crane and the average response time of all types of cranes, thus obtaining the crane's response characteristics. The standard response characteristics are the ratio of the response time of any type of crane to the average response time of all types of cranes. These standard characteristics are used as crane feature templates. The system obtains the crane's true characteristics by analyzing real-world crane data, including the crane's actual response time. This data is then used to derive the crane's true response characteristics, defined as the ratio of the crane's actual response time to the average response time of all crane types. This data is stored in historical data, timestamped, and includes the crane model and serial number. The crane serial number is assigned based on the crane model, creating a historical database of the crane's true characteristics. Furthermore, the system collects real-world crane operation data through situational awareness and compares it with the standard characteristics used in the factory. This dynamically captures performance deviations caused by wear and environmental changes, while also filtering historical operation data based on cargo characteristics. Compared to traditional control methods relying on preset parameters, this system can match the actual capabilities of the equipment with task requirements in real time, avoiding efficiency fluctuations caused by fixed parameters, effectively reducing the risk of task execution anomalies, meeting the precision and flexibility requirements of modern industry for crane operations, and adapting to different types of cargo operation scenarios.

[0012] Furthermore, in step S2, the historical database is retrieved, and tasks for the same type of goods are filtered to obtain X cranes performing the tasks. The historical data of the x-th crane is analyzed, where x = 1, 2, ..., X, and the number of tasks performed by the x-th crane is Y. Based on the historical data, the tasks are divided into M task stages, where the m-th task stage specifies N path points, m = 1, 2, ..., M. The y-th task is analyzed to obtain the completion degree A of the m-th stage in the y-th task. m The execution completion degree of the m-th stage is represented as follows: In the m-th stage, if the endpoint of the crane's task is not a predetermined endpoint, the execution completion degree of the m-th stage is 0; if the endpoint of the crane's task is a predetermined endpoint, the execution completion degree of the m-th stage is the ratio of the number of path points traversed by the crane through situational awareness to N, thereby obtaining the execution completion degree of the M stages {A1, A2, ..., A...}. m ,…,A M The system records crane numbers with completion information in a historical database. By searching the historical database, tasks targeting the same type of cargo are filtered to eliminate interference from different cargo characteristics, ensuring consistency in the analyzed tasks. This establishes a precise benchmark for subsequent assessments of equipment adaptability to specific cargo tasks, avoiding evaluation biases caused by traditional cross-type data mixing and improving the targeting of data analysis. The system breaks down tasks into multiple stages, assigning waypoints to each stage. This transforms the complex overall task into decomposable and quantifiable units. By judging whether the end point of each stage meets preset requirements and combining the actual passage of waypoints, the system calculates the completion rate, accurately locating the execution quality of each stage. This changes the traditional extensive control model that only focuses on the overall task result, achieving refined control from macro-level tasks to micro-level processes. The system stores completion data with crane identification information in a historical database, constructing a correlation file between equipment identity and the performance of each stage. This provides structured data support for tracing the root causes of abnormal stages and analyzing the impact of equipment characteristics on stages, promoting the transformation of data from simple collection to in-depth application, forming a closed loop of data application, and laying the foundation for the scientific nature of the overall control method.

[0013] Furthermore, in step S3, a threshold for the completion degree of each stage is set, specifying the execution degree B of the m-th stage. m When the completion rate A of the m-th step is... m If the execution completion rate is less than the threshold, the m-th step is considered to be abnormal, and the execution level of the m-th step is set to B. m =0; when the completion degree A of the m-th step is 0. m If the execution completion threshold is greater than or equal to the threshold value, the m-th step is considered to have executed normally, and the execution level of the m-th step is set to B. m=1, thus obtaining the execution degree of M steps in the y-th task {B1,B2,…,B m ,…,B M};

[0014] For the y-th task, the actual characteristics of the crane {C1,C2,…,C} are invoked. p ,…,C P}, invoke the standard features of the crane {D1,D2,…,D p ,…,D P}, where P represents the number of features of the crane, C p Let D represent the p-th real feature of the crane in the y-th task. p Let the p-th standard characteristic of the crane be represented, and then calculate the deviation rate E of the p-th characteristic of the crane in the y-th task. p E p =|C p -D p / D p |, set a deviation threshold for the p-th feature, and define the state G of the p-th feature. p When the p-th characteristic deviation rate E p When the deviation value is greater than the threshold value of the p-th feature, the p-th feature is judged to be abnormal, and the state of the p-th feature is set to G. p =0; when the p-th characteristic deviation rate E p When the deviation threshold of the p-th feature is less than or equal to the threshold value of the p-th feature, the p-th feature is judged to be normal, and the status of the p-th feature is set to G. p =1, thus obtaining P feature states {G1,G2,…,G} for the y-th task. p ,…,G P};

[0015] Analyze the influence coefficient F of the p-th feature on the m-th stage. p The F p This means that if the number of tasks in task Y that satisfy the condition that the execution level of the m-th stage is 0 is 0, then F... p =0; If the number of tasks in task Y that satisfy the condition that the execution degree of the m-th stage is 0 is not 0, F p Let Fm be the ratio of the number of tasks in task Y where the execution level of the m-th stage is 0 and the p-th feature condition is 0 to the number of tasks where the execution level of the m-th stage is 0. Then, the influence coefficients {F1, F2, ..., Fp} of the P features on the m-th stage can be obtained. p ,…,F P};

[0016] By setting thresholds for the completion of each stage, the normal and abnormal states of stage execution are clearly distinguished, solving the problem of vague judgments on stage performance in traditional methods and providing a clear basis for accurately locating problematic stages in a task. Comparing real and standard characteristics, and combining this with characteristic deviation thresholds to judge characteristic anomalies, allows for precise capture of equipment performance deviations, a clear understanding of the true state of the equipment, and overcomes the shortcomings of traditional methods that struggle to quantify characteristic anomalies. Analyzing the impact coefficients of characteristics on stages and establishing a correlation between the two breaks down the previous analytical model that separated equipment characteristics from operational stages, significantly improving the efficiency of anomaly tracing and providing a clear direction for developing targeted improvement measures.

[0017] Furthermore, in step S4, the impact of the execution level of the m-th stage on the task completion level is analyzed. Based on the crane's final task requirements, a task completion level is set. These final task requirements can be set as energy consumption conditions, such as the ratio of ineffective work to total consumed work not exceeding 50%. When the task fails to meet the final task requirements, the task is judged as failed, and the task completion level is set to 0. When the task meets the final task requirements, the task is judged as successful, and the task completion level is set to 1. This yields the impact H of the execution level of the m-th stage on the task completion level. m The influence of the execution level of the m-th step on the task completion level is represented by the ratio of the number of tasks in task Y with a task completion level of 0 and a execution level of 0 in the m-th step to the number of tasks with an influence level of 0 on task completion. This yields the correlation parameters {J1, J2, ..., J...} of the M steps on the task completion level. m ,…,J M}, where J m This represents the correlation coefficient between the m-th step and the task completion level. By clarifying the task completion standards based on the final task requirements, it clearly distinguishes between success and failure states, solving the problem of ambiguity in traditional task completion assessments and providing a clear basis for task results. Simultaneously, by analyzing the correlation between step execution anomalies and task failures, it identifies the steps that influence task success or failure, moving away from the previous indiscriminate assessment of the importance of each step and clarifying the criticality of different steps. Furthermore, the generated correlation coefficient between steps and task completion provides precise direction for prioritizing and controlling key steps and optimizing task processes, helping to improve the overall task success rate from the perspective of step management.

[0018] Furthermore, in step S5, upon receiving a request to execute a task for any type of cargo, the historical database is retrieved, and tasks for the same type of cargo are filtered. The suitability of the x-th crane for the task is analyzed, and the anomaly rate of the P features is {R1, R2, ..., R...}. p ,…,R P}, where R pLet Fx be the anomaly rate of the p-th feature of the x-th crane, where the anomaly rate of the p-th feature represents the ratio of the number of tasks with a feature status of 0 in Y tasks to Yi. Let Fi be the influence coefficient {F1, F2, ..., Fi} of the p-th feature on the m-th stage. p ,…,F P}, calculate the prediction anomaly rate w of the m-th stage. m :

[0019] ;

[0020] Substituting each value into m = 1, 2, ..., M, we obtain the prediction anomaly rates {w1, w2, ..., w} for the M stages. m ,…,w M}, call the correlation parameters {J1, J2, ..., J} of M stages to determine the degree of task completion. m ,…,J M} Calculate the suitability W of the x-th crane for the task. x :

[0021] ;

[0022] Then, the X cranes are analyzed to obtain the suitability of the X cranes for the task {W1, W2, ..., W...}. x ,…,W X The system sorts X cranes from highest to lowest suitability, selecting x0 cranes as candidate cranes, where x0 is the predetermined number of candidate cranes. Suitability is assessed on these x0 candidate cranes using a set suitability threshold. If the suitability of any candidate crane exceeds the threshold, it is included in the pre-selection list; otherwise, it is ignored. Historical database searches are used to filter task data for similar goods, ensuring the suitability analysis revolves around task characteristics and avoiding cross-type data interference, thus addressing the lack of specificity in traditional allocation. By combining abnormal crane characteristics with their impact on various stages, the system predicts the risk of abnormalities in each stage and calculates the suitability based on the relationship between stages and task completion. This data-driven approach, rather than relying solely on experience, significantly improves accuracy. By sorting cranes by suitability and using thresholds to filter candidate cranes, poorly suited equipment is eliminated, avoiding task risks caused by blind allocation. This lays the foundation for accurate equipment selection and task execution, ultimately improving task completion success rates.

[0023] Furthermore, in step S6, after obtaining Q pre-selected cranes, the working status of the Q pre-selected cranes is collected based on situational awareness to obtain the remaining working time {T1,T2,…,T} of the Q pre-selected cranes. q ,…,T Q}, where T qLet q be the remaining working time of the pre-selected crane. The remaining working time of the q-th pre-selected crane is defined as follows: when the q-th pre-selected crane has an assigned task, the remaining working time of the q-th pre-selected crane is the estimated time for the pre-selected crane to complete the assigned task; otherwise, the remaining working time of the q-th pre-selected crane is 0. When there is a pre-selected crane with a remaining working time of 0, the crane with the highest suitability among the pre-selected cranes with a remaining working time of 0 is selected to receive the task; when there is no pre-selected crane with a remaining working time of 0, the Q pre-selected cranes are prioritized, and the priority ranking score of the q-th pre-selected crane is: the ratio of the suitability of the q-th pre-selected crane to the remaining working time of the pre-selected crane, divided by the sum of the remaining working times of the Q pre-selected cranes, and then the crane with the highest priority ranking score is selected to receive the task. By collecting the remaining working time of the pre-selected cranes through situational awareness, the real-time busy / idle status of the equipment can be accurately grasped, avoiding the allocation deviation caused by relying solely on subjective judgment of equipment idleness in traditional methods. When available equipment, priority is given to the most compatible devices to ensure tasks are performed by highly compatible equipment and guarantee task quality. When no equipment is available, a priority score is calculated based on compatibility and remaining time to avoid delays caused by simply waiting for available equipment or selecting poorly compatible equipment that could negatively impact completion. This approach balances equipment compatibility with operational timeliness, significantly improving the rationality of task allocation, ensuring efficient task progress, and optimizing the utilization efficiency of port crane resources.

[0024] The situational awareness-based container crane control system includes: a crane data database module, a historical task process analysis module, a process state deviation analysis module, a process task correlation analysis module, a real-time crane pre-selection module, and a crane task allocation module.

[0025] The crane data database module is used to collect standard data of cranes, and after preprocessing and normalization, establish a historical database of crane characteristics.

[0026] The historical task segment analysis module is used to retrieve historical databases, filter similar cargo tasks to obtain cranes that perform tasks, analyze historical data, break down task segments, and analyze the completion rate of task segments.

[0027] The process state deviation analysis module is used to determine the process execution status, analyze the characteristic deviation status, and then analyze the degree of influence of the characteristic on the process based on the characteristic deviation status.

[0028] The task correlation analysis module is used to analyze the impact of the execution status of the process on the task completion status, and to obtain the correlation coefficient between the execution status of the process and the task completion status by comprehensively considering the degree of task completion.

[0029] The real-time crane pre-selection module is used to receive tasks in real time, retrieve historical data, analyze the adaptability of cranes to tasks, and obtain pre-selected cranes.

[0030] The crane task allocation is used to comprehensively consider the remaining working time of the pre-selected cranes and select a crane to receive the lifting task.

[0031] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: By collecting real-time crane operating data through situational awareness and comparing it with standard factory characteristics, it dynamically captures performance deviations caused by wear and environmental changes, while simultaneously filtering historical operation data based on cargo characteristics. Compared to traditional control methods that rely on preset parameters, it can match the actual capabilities of the equipment with task requirements in real time, avoiding efficiency fluctuations caused by fixed parameters, effectively reducing the risk of task execution anomalies, meeting the requirements of modern industry for the precision and flexibility of crane operations, and adapting to different types of cargo operation scenarios;

[0032] By calculating the influence coefficient of features on processes and the correlation coefficient between processes and task completion, a multi-layered correlation system is established between equipment features, operational processes, and task completion. When execution deviations or task failures occur, the root cause of the anomaly (abnormal equipment features or operational problems in processes) can be accurately located, replacing the traditional single-phenomenon judgment mode. This not only improves the efficiency of anomaly investigation but also enables targeted preventive measures to be developed, reducing the recurrence of similar anomalies and enhancing overall operational stability.

[0033] The task allocation process no longer relies solely on equipment idle status. Instead, it combines historical crane performance and characteristic stability calculations to determine suitability, and then prioritizes tasks based on remaining working time. This avoids assigning tasks to idle equipment with poor suitability, improving the efficiency of single task execution, reducing the probability of failures due to equipment and task mismatch, ensuring smooth overall port operations, and further optimizing resource utilization efficiency. Attached Figure Description

[0034] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0035] Figure 1 This is a structural diagram of the container crane control system based on situational awareness of the present invention;

[0036] Figure 2 This is a flowchart of the container crane control method based on situational awareness according to the present invention. Detailed Implementation

[0037] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0038] Please see Figure 1 and Figure 2 This invention provides a technical solution: a container crane control method based on situational awareness, comprising the following steps:

[0039] S1. Collect standard data of cranes, and after preprocessing and normalization, establish a historical database of crane characteristics;

[0040] S2. Search the historical database, filter similar cargo tasks to obtain the cranes that perform the tasks, analyze the historical data, break down the task steps, and analyze the completion rate of the task steps;

[0041] S3. Determine the execution status of the process, analyze the characteristic deviation, and then analyze the degree of influence of the characteristic on the process based on the characteristic deviation.

[0042] S4. Analyze the impact of the execution status of each step on the task completion status, and comprehensively consider the degree of task completion to obtain the correlation coefficient between the execution status of each step and the task completion status.

[0043] S5. After receiving the task in real time, retrieve historical data, analyze the adaptability of the crane to the task, and obtain the pre-selected crane.

[0044] S6. Taking into account the remaining working time of the pre-selected cranes, select a crane to accept the lifting task.

[0045] Furthermore, in step S1, standard data of the crane is collected. This standard data is the data specified in the crane's factory specifications. Based on situational awareness, real-world data of the crane is collected. This real-world data is the data obtained during crane operation based on situational awareness. After preprocessing and normalization, standard features of the crane are obtained through the standard data. These standard features are used as crane feature templates. Real-world features of the crane are obtained through the real-world data, stored in historical data, and stamped with a timestamp, crane model, and crane number. The crane number is assigned based on the crane model, thus obtaining a historical database of the crane's real-world features. Real-world operating data of the crane is collected through situational awareness and compared with the factory standard features to dynamically capture performance deviations caused by wear and environmental changes. Simultaneously, historical operation data is filtered based on cargo characteristics. Compared to traditional control methods that rely on preset parameters, this approach can match the actual capabilities of the equipment with task requirements in real time, avoiding efficiency fluctuations caused by fixed parameters, effectively reducing the risk of abnormal task execution, meeting the requirements of modern industry for the accuracy and flexibility of crane operations, and adapting to different types of cargo operation scenarios.

[0046] In step S2, the historical database is retrieved, and tasks for the same type of goods are filtered to obtain X cranes performing the tasks. The historical data of the x-th crane is analyzed, where x = 1, 2, ..., X, and the number of tasks performed by the x-th crane is Y. Based on the historical data, the tasks are divided into M task stages, where the m-th task stage has N path points, m = 1, 2, ..., M. The completion degree A of the m-th stage in the y-th task is analyzed. m The execution completion degree of the m-th stage is represented as follows: In the m-th stage, if the endpoint of the crane's task is not a predetermined endpoint, the execution completion degree of the m-th stage is 0; if the endpoint of the crane's task is a predetermined endpoint, the execution completion degree of the m-th stage is the ratio of the number of path points traversed by the crane through situational awareness to N, thereby obtaining the execution completion degree of the M stages {A1, A2, ..., A...}. m ,…,A MThe system records crane numbers with completion information in a historical database. By searching the historical database, tasks targeting the same type of cargo are filtered to eliminate interference from different cargo characteristics, ensuring consistency in the analyzed tasks. This establishes a precise benchmark for subsequent assessments of equipment adaptability to specific cargo tasks, avoiding evaluation biases caused by traditional cross-type data mixing and improving the targeting of data analysis. The system breaks down tasks into multiple stages, assigning waypoints to each stage. This transforms the complex overall task into decomposable and quantifiable units. By judging whether the end point of each stage meets preset requirements and combining the actual passage of waypoints, the system calculates the completion rate, accurately locating the execution quality of each stage. This changes the traditional extensive control model that only focuses on the overall task result, achieving refined control from macro-level tasks to micro-level processes. The system stores completion data with crane identification information in a historical database, constructing a correlation file between equipment identity and the performance of each stage. This provides structured data support for tracing the root causes of abnormal stages and analyzing the impact of equipment characteristics on stages, promoting the transformation of data from simple collection to in-depth application, forming a closed loop of data application, and laying the foundation for the scientific nature of the overall control method.

[0047] In step S3, a threshold for the completion degree of each stage is set, specifying the execution degree B of the m-th stage. m When the completion rate A of the m-th step is... m If the execution completion rate is less than the threshold, the m-th step is considered to be abnormal, and the execution level of the m-th step is set to B. m =0; when the completion degree A of the m-th step is 0. m If the execution completion threshold is greater than or equal to the threshold value, the m-th step is considered to have executed normally, and the execution level of the m-th step is set to B. m =1, thus obtaining the execution degree of M steps in the y-th task {B1,B2,…,B m ,…,B M};

[0048] For the y-th task, the actual characteristics of the crane {C1,C2,…,C} are invoked. p ,…,C P}, invoke the standard features of the crane {D1,D2,…,D p ,…,D P}, where P represents the number of features of the crane, C p Let D represent the p-th real feature of the crane in the y-th task. p Let the p-th standard characteristic of the crane be represented, and then calculate the deviation rate E of the p-th characteristic of the crane in the y-th task. p E p =|C p -D p / D p|, set a deviation threshold for the p-th feature, and define the state G of the p-th feature. p When the p-th characteristic deviation rate E p When the deviation value is greater than the threshold value of the p-th feature, the p-th feature is judged to be abnormal, and the state of the p-th feature is set to G. p =0; when the p-th characteristic deviation rate E p When the deviation threshold of the p-th feature is less than or equal to the threshold value of the p-th feature, the p-th feature is judged to be normal, and the status of the p-th feature is set to G. p =1, thus obtaining P feature states {G1,G2,…,G} for the y-th task. p ,…,G P};

[0049] Analyze the influence coefficient F of the p-th feature on the m-th stage. p The F p This means that if the number of tasks in task Y that satisfy the condition that the execution level of the m-th stage is 0 is 0, then F... p =0; If the number of tasks in task Y that satisfy the condition that the execution degree of the m-th stage is 0 is not 0, F p Let Fm be the ratio of the number of tasks in task Y where the execution level of the m-th stage is 0 and the p-th feature condition is 0 to the number of tasks where the execution level of the m-th stage is 0. Then, the influence coefficients {F1, F2, ..., Fp} of the P features on the m-th stage can be obtained. p ,…,F P};

[0050] By setting thresholds for the completion of each stage, the normal and abnormal states of stage execution are clearly distinguished, solving the problem of vague judgments on stage performance in traditional methods and providing a clear basis for accurately locating problematic stages in a task. Comparing real and standard characteristics, and combining this with characteristic deviation thresholds to judge characteristic anomalies, allows for precise capture of equipment performance deviations, a clear understanding of the true state of the equipment, and overcomes the shortcomings of traditional methods that struggle to quantify characteristic anomalies. Analyzing the impact coefficients of characteristics on stages and establishing a correlation between the two breaks down the previous analytical model that separated equipment characteristics from operational stages, significantly improving the efficiency of anomaly tracing and providing a clear direction for developing targeted improvement measures.

[0051] In step S4, the impact of the execution level of the m-th stage on the task completion level is analyzed. Based on the crane's final task requirements, a task completion level is set. These final task requirements can be set as energy consumption conditions, such as the ratio of ineffective work to total consumed work not exceeding 50%. When the task fails to meet the final task requirements, the task is considered a failure, and the task completion level is set to 0. When the task meets the final task requirements, the task is considered a success, and the task completion level is set to 1. This yields the impact H of the execution level of the m-th stage on the task completion level. mThe influence of the execution level of the m-th step on the task completion level is represented by the ratio of the number of tasks in task Y with a task completion level of 0 and a execution level of 0 in the m-th step to the number of tasks with an influence level of 0 on task completion. This yields the correlation parameters {J1, J2, ..., J...} of the M steps on the task completion level. m ,…,J M}, where J m This represents the correlation coefficient between the m-th step and the task completion level. By clarifying the task completion standards based on the final task requirements, it clearly distinguishes between success and failure states, solving the problem of ambiguity in traditional task completion assessments and providing a clear basis for task results. Simultaneously, by analyzing the correlation between step execution anomalies and task failures, it accurately identifies the steps affecting task success or failure, breaking away from the previous crude model of indiscriminately judging the importance of each step and clarifying the criticality of different steps. Furthermore, the generated correlation coefficient between steps and task completion provides precise direction for subsequent priority management of key steps and optimization of task processes, helping to improve the overall task success rate from the perspective of step management.

[0052] In step S5, when a task is received to execute for any type of cargo, the historical database is retrieved, and tasks for the same type of cargo are filtered. The suitability of the x-th crane for the task is analyzed, and the anomaly rate of P features is {R1, R2, ..., R...}. p ,…,R P}, where R p Let Fx be the anomaly rate of the p-th feature of the x-th crane, where the anomaly rate of the p-th feature represents the ratio of the number of tasks with a feature status of 0 in Y tasks to Yi. Let Fi be the influence coefficient {F1, F2, ..., Fi} of the p-th feature on the m-th stage. p ,…,F P}, calculate the prediction anomaly rate w of the m-th stage. m :

[0053] ;

[0054] Substituting each value into m = 1, 2, ..., M, we obtain the prediction anomaly rates {w1, w2, ..., w} for the M stages. m ,…,w M}, call the correlation parameters {J1, J2, ..., J} of M stages to determine the degree of task completion. m ,…,J M} Calculate the suitability W of the x-th crane for the task. x :

[0055] ;

[0056] Then, the X cranes are analyzed to obtain the suitability of the X cranes for the task {W1, W2, ..., W...}. x ,…,W X The system sorts X cranes from highest to lowest suitability, selecting x0 cranes as candidate cranes, where x0 is the predetermined number of candidate cranes. Suitability is assessed on these x0 candidate cranes using a set suitability threshold. If the suitability of any candidate crane exceeds the threshold, it is included in the pre-selection list; otherwise, it is ignored. Historical database searches are used to filter task data for similar goods, ensuring the suitability analysis revolves around task characteristics and avoiding cross-type data interference, thus addressing the lack of specificity in traditional allocation. By combining abnormal crane characteristics with their impact on various stages, the system predicts the risk of abnormalities in each stage and calculates the suitability based on the relationship between stages and task completion. This data-driven approach, rather than relying solely on experience, significantly improves accuracy. By sorting cranes by suitability and using thresholds to filter candidate cranes, poorly suited equipment is eliminated, avoiding task risks caused by blind allocation. This lays the foundation for accurate equipment selection and task execution, ultimately improving task completion success rates.

[0057] In step S6, after obtaining Q pre-selected cranes, the working status of the Q pre-selected cranes is collected based on situational awareness, and the remaining working time {T1,T2,…,T} of the Q pre-selected cranes is obtained. q ,…,T Q}, where T qLet q be the remaining working time of the pre-selected crane. The remaining working time of the q-th pre-selected crane is defined as follows: when the q-th pre-selected crane has an assigned task, the remaining working time of the q-th pre-selected crane is the estimated time for the pre-selected crane to complete the assigned task; otherwise, the remaining working time of the q-th pre-selected crane is 0. When there is a pre-selected crane with a remaining working time of 0, the crane with the highest suitability among the pre-selected cranes with a remaining working time of 0 is selected to receive the task; when there is no pre-selected crane with a remaining working time of 0, the Q pre-selected cranes are prioritized, and the priority ranking score of the q-th pre-selected crane is: the ratio of the suitability of the q-th pre-selected crane to the remaining working time of the pre-selected crane, divided by the sum of the remaining working times of the Q pre-selected cranes, and then the crane with the highest priority ranking score is selected to receive the task. By collecting the remaining working time of the pre-selected cranes through situational awareness, the real-time busy / idle status of the equipment can be accurately grasped, avoiding the allocation deviation caused by relying solely on subjective judgment of equipment idleness in traditional methods. When available equipment, priority is given to the most compatible devices to ensure tasks are performed by highly compatible equipment and guarantee task quality. When no equipment is available, a priority score is calculated based on compatibility and remaining time to avoid delays caused by simply waiting for available equipment or selecting poorly compatible equipment that could negatively impact completion. This approach balances equipment compatibility with operational timeliness, significantly improving the rationality of task allocation, ensuring efficient task progress, and optimizing the utilization efficiency of port crane resources.

[0058] The situational awareness-based container crane control system includes: a crane data database module, a historical task process analysis module, a process state deviation analysis module, a process task correlation analysis module, a real-time crane pre-selection module, and a crane task allocation module.

[0059] The crane data database module is used to collect standard crane data, and after preprocessing and normalization, establish a historical database of crane characteristics.

[0060] The historical task segment analysis module is used to retrieve historical databases, filter similar cargo tasks to obtain cranes that perform tasks, analyze historical data, break down task segments, and analyze the completion rate of task segments.

[0061] The process status deviation analysis module is used to determine the execution status of a process, analyze the characteristic deviation, and then analyze the degree of influence of the characteristic on the process based on the characteristic deviation.

[0062] The task correlation analysis module is used to analyze the impact of the execution status of a step on the completion of a task, and to obtain the correlation coefficient between the execution status of a step and the completion of a task by comprehensively considering the degree of task completion.

[0063] The real-time crane pre-selection module is used to receive tasks in real time, retrieve historical data, analyze the adaptability of cranes to tasks, and obtain pre-selected cranes.

[0064] Crane task allocation is used to comprehensively consider the remaining working time of the pre-selected cranes and select the cranes to receive the lifting tasks.

[0065] Example 1: First, data collection and database construction were carried out. Standard data specified in the crane's manufacturer's specifications were collected, and real-world operational data was acquired using situational awareness technology during crane operation. After preprocessing and normalizing both types of data, standard features were extracted from the standard data and used as feature templates, while real-world features were extracted from the real-world data. When storing real-world features, the collection time, crane model, and unique serial number were simultaneously labeled. The unique serial number was assigned separately for each crane model, ultimately constructing a historical database containing the real-world features of the cranes.

[0066] Next, historical task data is analyzed. Operational task records for the same type of goods are retrieved from the historical database, and cranes that have performed these tasks are identified. The historical operational data of these cranes is analyzed, dividing each task into multiple operational stages, each with pre-defined waypoints. The completion status of each stage is assessed: if the endpoint of a stage differs from the pre-defined endpoint, the completion rate of that stage is considered insufficient; if the endpoint matches the pre-defined endpoint, the actual waypoints obtained from situational awareness are compared with the pre-defined waypoints for that stage to determine its completion rate. This completion information is then labeled with the corresponding crane number and re-stored in the historical database.

[0067] Subsequently, an anomaly detection mechanism was established. Standards were set to judge the execution status of each stage. If the completion rate of a stage did not meet the standard, the stage was considered an anomaly; if it met the standard, execution was considered normal. Simultaneously, by comparing the actual characteristics of the crane with standard characteristics, standards were set to judge whether a characteristic was normal. If the difference exceeded the standard, the characteristic was considered anomaly; if it did not exceed the standard, the characteristic was considered normal. In this way, the execution status of each stage and the status of each characteristic in each task were determined separately.

[0068] Next, we analyze the impact of features on the process. We review past work records. If a process has never experienced execution anomalies in past tasks, it indicates that the relevant features have a relatively small impact on that process. If a process has experienced execution anomalies, we count the number of times the corresponding feature anomalies occurred in those anomalies, and use these statistical results to determine the degree of impact of each feature on that process.

[0069] The impact of each step on the overall task completion is then analyzed. The task's completion status is determined based on its final requirements; failure indicates task failure, while success indicates success. Past tasks are analyzed for instances where tasks failed and corresponding steps exhibited abnormal execution; these instances are then used to determine the correlation between each step and the task completion outcome.

[0070] Finally, task adaptation and crane selection are performed. When a new task involving similar goods is received, the historical database is searched again to analyze the adaptability of each crane to the task. First, any anomalies in the characteristics of each crane are examined. Combining this with the previously determined impact of the characteristics on the process and the relationship between the process and the task, the adaptability of each crane is calculated. After sorting by adaptability, candidate cranes are selected. Then, it is determined whether the adaptability of the candidate cranes meets the set requirements; those that meet the requirements are pre-selected cranes.

[0071] Check the remaining working hours of these pre-selected cranes. If there are cranes that do not currently have any assigned tasks, prioritize the one with the highest suitability to receive the new task. If all pre-selected cranes have assigned tasks, calculate the priority of each crane based on the ratio of each crane's suitability to its remaining working hours, combined with the overall situation of the remaining working hours of all pre-selected cranes. Finally, select the crane with the highest priority to execute the new task.

[0072] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary sensing device embodiments described above, and that the invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A container crane control method based on situation awareness, characterized by: The method includes the following steps: S1. Collect standard data of cranes, and after preprocessing and normalization, establish a historical database of crane characteristics; S2. Search the historical database, filter similar cargo tasks to obtain the cranes that perform the tasks, analyze the historical data, break down the task steps, and analyze the completion rate of the task steps; S3. Determine the execution status of the process, analyze the characteristic deviation, and then analyze the degree of influence of the characteristic on the process based on the characteristic deviation. S4. Analyze the impact of the execution status of each step on the task completion status, and comprehensively consider the degree of task completion to obtain the correlation coefficient between the execution status of each step and the task completion status. S5. After receiving the task in real time, retrieve historical data, analyze the adaptability of the crane to the task, and obtain the pre-selected crane. S6. Taking into account the remaining working time of the pre-selected cranes, select a crane to accept the lifting task. In step S1, standard data of the crane is collected. The standard data is the data specified in the crane's factory specifications. Based on situational awareness, real data of the crane is collected. The real data is the data obtained by the crane during operation based on situational awareness. After preprocessing and normalization, the standard features of the crane are obtained through the standard data of the crane. The standard features are used as the crane feature template. The real features of the crane are obtained through the real data of the crane. The real data is stored in historical data and stamped with timestamps, crane model and crane number. The crane number is numbered according to the crane model. In this way, a historical database of the real features of the crane is obtained.

2. The situational awareness based container crane control method of claim 1, wherein: In step S2, the historical database is retrieved, and tasks for the same type of goods are filtered to obtain X cranes performing the tasks. The historical data of the x-th crane is analyzed, where x = 1, 2, ..., X, and the number of tasks performed by the x-th crane is Y. Based on the historical data, the tasks are divided into M task stages, where the m-th task stage has N path points, m = 1, 2, ..., M. The completion degree A of the m-th stage in the y-th task is analyzed. m The execution completion degree of the m-th stage is represented as follows: In the m-th stage, if the endpoint of the crane's task is not a predetermined endpoint, the execution completion degree of the m-th stage is 0; if the endpoint of the crane's task is a predetermined endpoint, the execution completion degree of the m-th stage is the ratio of the number of path points traversed by the crane through situational awareness to N, thereby obtaining the execution completion degree of the M stages {A1, A2, ..., A...}. m ,…,A M The crane number marked with the completion status will be stored in the historical database.

3. The situationally aware container crane control method of claim 2, wherein: In step S3, a threshold for the completion degree of each stage is set, specifying the execution degree B of the m-th stage. m When the completion rate A of the m-th step is... m If the execution completion rate is less than the threshold, the m-th step is considered to be abnormal, and the execution level of the m-th step is set to B. m =0; when the completion degree A of the m-th step is 0. m If the execution completion threshold is greater than or equal to the threshold value, the m-th step is considered to have executed normally, and the execution level of the m-th step is set to B. m =1, thus obtaining the execution degree of M steps in the y-th task {B1,B2,…,B m ,…,B M }; For the y-th task, the actual characteristics of the crane {C1,C2,…,C} are invoked. p ,…,C P }, invoke the standard features of the crane {D1,D2,…,D p ,…,D P }, where P represents the number of features of the crane, C p Let D represent the p-th real feature of the crane in the y-th task. p Let the p-th standard characteristic of the crane be represented, and then calculate the deviation rate E of the p-th characteristic of the crane in the y-th task. p E p =|C p -D p / D p |, set a deviation threshold for the p-th feature, and define the state G of the p-th feature. p When the p-th characteristic deviation rate E p When the deviation value is greater than the threshold value of the p-th feature, the p-th feature is judged to be abnormal, and the state of the p-th feature is set to G. p =0; When the p-th characteristic deviation rate E p When the deviation threshold of the p-th feature is less than or equal to the threshold value of the p-th feature, the p-th feature is judged to be normal, and the status of the p-th feature is set to G. p =1, thus obtaining P feature states {G1,G2,…,G} for the y-th task. p ,…,G P } 4. The situationally aware container crane control method of claim 3, wherein: Analyze the influence coefficient F of the p-th feature on the m-th stage. p The F p This means that if the number of tasks in task Y that satisfy the condition that the execution level of the m-th stage is 0 is 0, then F... p =0; If the number of tasks in task Y that satisfy the condition that the execution degree of the m-th stage is 0 is not 0, F p Let Fm be the ratio of the number of tasks in task Y where the execution level of the m-th stage is 0 and the p-th feature condition is 0 to the number of tasks where the execution level of the m-th stage is 0. Then, the influence coefficients {F1, F2, ..., Fp} of the P features on the m-th stage can be obtained. p ,…,F P } 5. The container crane control method based on situational awareness according to claim 4, characterized in that: In step S4, the impact of the execution level of the m-th stage on the task completion level is analyzed. Based on the final task requirements of the crane, a task completion level is set. When the task fails to meet the final task requirements, the task is judged as failed, and the task completion level is set to 0; when the task meets the final task requirements, the task is judged as successful, and the task completion level is set to 1. Thus, the impact H of the execution level of the m-th stage on the task completion level is obtained. m The influence of the execution level of the m-th step on the task completion level is represented by the ratio of the number of tasks in task Y with a task completion level of 0 and a execution level of 0 in the m-th step to the number of tasks with an influence level of 0 on task completion. This yields the correlation parameters {J1, J2, ..., J...} of the M steps on the task completion level. m ,…,J M }, where J m This represents the correlation coefficient between the m-th step and the degree of task completion.

6. The situationally aware container crane control method of claim 5, wherein: In step S5, when a task is received to execute for any type of cargo, the historical database is retrieved, and tasks for the same type of cargo are filtered. The suitability of the x-th crane for the task is analyzed, and the anomaly rate of P features is {R1, R2, ..., R...}. p ,…,R P }, where R p Let Fx be the anomaly rate of the p-th feature of the x-th crane, where the anomaly rate of the p-th feature represents the ratio of the number of tasks with a feature status of 0 in Y tasks to Yi. Let Fi be the influence coefficient {F1, F2, ..., Fi} of the p-th feature on the m-th stage. p ,…,F P }, calculate the prediction anomaly rate w of the m-th stage. m : ； Substituting each value into m = 1, 2, ..., M, we obtain the prediction anomaly rates {w1, w2, ..., w} for the M stages. m ,…,w M }, call the correlation parameters {J1, J2, ..., J} of M stages to determine the degree of task completion. m ,…,J M } Calculate the suitability W of the x-th crane for the task. x : ； Then, the X cranes are analyzed to obtain the suitability of the X cranes for the task {W1, W2, ..., W...}. x ,…,W X The X cranes are sorted from highest to lowest suitability. The x0 cranes with the highest suitability are selected as candidate cranes, where x0 is the predetermined number of candidate cranes. The suitability of the x0 candidate cranes is judged, and a suitability threshold is set. If the suitability of any candidate crane is greater than the predetermined suitability threshold, the candidate crane is selected into the pre-selection cranes; otherwise, the candidate crane is ignored.

7. The situationally aware container crane control method of claim 6, wherein: In step S6, after obtaining Q pre-selected cranes, the working status of the Q pre-selected cranes is collected based on situational awareness, and the remaining working time {T1,T2,…,T} of the Q pre-selected cranes is obtained. q ,…,T Q }, where T q Let q be the remaining working time of the qth pre-selected crane. The remaining working time of the qth pre-selected crane means: when the qth pre-selected crane has an assigned task, the remaining working time of the qth pre-selected crane is the estimated time for the pre-selected crane to complete the assigned task; otherwise, the remaining working time of the qth pre-selected crane is 0. When there is a pre-selected crane with a remaining working time of 0, the crane with the highest suitability is selected from the pre-selected cranes with a remaining working time of 0 to receive the task. When there are no pre-selected cranes with 0 remaining working time, the Q pre-selected cranes are prioritized. The priority ranking score of the q-th pre-selected crane is calculated by dividing the ratio of the suitability of the q-th pre-selected crane to the remaining working time of the pre-selected crane by the sum of the remaining working times of the Q pre-selected cranes. The crane with the highest priority ranking score is then selected to receive the task.

8. A situational awareness-based container crane control system, wherein the system is applied to the situational awareness-based container crane control method according to any one of claims 1-7, characterized in that: The system includes: a crane data database module, a historical task process analysis module, a process status deviation analysis module, a process task correlation analysis module, a real-time crane pre-selection module, and a crane task allocation module; The crane data database module is used to collect standard data of cranes, and after preprocessing and normalization, establish a historical database of crane characteristics. The historical task segment analysis module is used to retrieve historical databases, filter similar cargo tasks to obtain cranes that perform tasks, analyze historical data, break down task segments, and analyze the completion rate of task segments. The process state deviation analysis module is used to determine the process execution status, analyze the characteristic deviation status, and then analyze the degree of influence of the characteristic on the process based on the characteristic deviation status. The task correlation analysis module is used to analyze the impact of the execution status of the stage on the task completion status, and to obtain the correlation coefficient between the execution status of the stage and the task completion status by comprehensively considering the degree of task completion. The real-time crane pre-selection module is used to receive tasks in real time, retrieve historical data, analyze the adaptability of cranes to tasks, and obtain pre-selected cranes. The crane task allocation is used for comprehensively considering the residual working time length of the preselected crane, and selecting the crane to receive the crane task.

Images