Real-time workload monitoring method and system based on sensor network

Abstract

The invention discloses a real-time workload monitoring method and system based on a sensor network. The method comprises the following steps: arranging various types of sensors in a ship; setting a unified basic sampling frequency for each sensor; transmitting the acquired data to a central data processing platform through a sensor network; the received sensor data is preprocessed on the central data processing platform; and analyzing the preprocessed data by using a first neural network model of a machine learning algorithm, and identifying the workload state of the crew. According to the invention, work and life conditions of sailors are comprehensively monitored through a sensor network, data can be transmitted in real time, and a basis is provided for decision making; and a ship manager is helped to timely obtain the health and working states of the sailors, so that resource allocation and working arrangement are optimized. And by monitoring the environmental factors, the manager can adjust the navigation strategy in time, and the negative influence on the sailors is reduced.

Description

Technical Field

[0001] The present invention belongs to the field of computer system engineering, and in particular relates to a real-time workload monitoring method and system based on a sensor network. Background Art

[0002] In modern marine transportation and shipping, real-time crew monitoring technology has become an essential component for ensuring safe, efficient, and sustainable operations. As ship sizes increase and navigation environments become more complex, real-time monitoring of crew workload, health status, and physiological responses has become increasingly important. The following is a detailed explanation of the background technology behind real-time crew monitoring.

[0003] Crew members face various challenges during voyages, including long working hours, intense physical exertion, mental stress, and complex environmental factors. These factors can lead to excessive workloads, which in turn can impact crew health and safety. Real-time monitoring of crew physiological status (such as heart rate, blood pressure, and fatigue levels) and environmental conditions (such as temperature, humidity, and noise) can help managers promptly identify potential risks and implement appropriate measures to safeguard crew health and safety.

[0004] Although some monitoring technologies have been applied to crew management, existing systems often have the following limitations:

[0005] Data silos: Many monitoring systems operate independently, lack effective data integration, and are unable to provide comprehensive real-time information.

[0006] Response delay: Traditional monitoring methods usually rely on periodic inspections and cannot provide real-time feedback, resulting in potential problems not being discovered in time.

[0007] Technical adaptability: The ship environment is complex and changeable, and the existing sensors and monitoring equipment are not adaptable enough, which may lead to inaccurate data collection. Summary of the Invention

[0008] In view of the above-mentioned defects in the prior art, the present invention provides a real-time workload monitoring method based on a sensor network, comprising the following steps:

[0009] Step S101: arranging multiple types of sensors in a ship, wherein the multiple types of sensors include physiological sensors, environmental sensors, and work task sensors;

[0010] Step S103: For each sensor, set a unified basic sampling frequency;

[0011] Step S105: Transmitting the collected data to a central data processing platform via the sensor network;

[0012] Step S107: pre-processing the received sensor data on the central data processing platform;

[0013] Step S109: Use the first neural network model of the machine learning algorithm to analyze the preprocessed data to identify the workload status of the crew.

[0014] The physiological sensors include a heart rate monitor and a blood oxygen saturation sensor, the environmental sensors include a temperature sensor, a humidity sensor and a noise sensor, and the work task sensors include a task completion time recorder and a work intensity evaluator.

[0015] Wherein, the step S105 includes:

[0016] Each sensor packages the real-time data through the producer mode and sends it to Kafka of the central processing platform;

[0017] The Kafka stores the received real-time data stream in a distributed log.

[0018] Wherein, the step S107 includes:

[0019] The Flink of the central processing platform processes each message sent from the Kafka and performs real-time analysis.

[0020] The first neural network model in step S109 is implemented by a multi-layer perceptron structure, specifically including an input layer, a hidden layer and an output layer, wherein the hidden layer adopts a ReLU activation function.

[0021] The first neural network model uses the following formula to evaluate the crew's real-time workload:

[0022] Among them, H i represents the real-time workload evaluation value of crew member i; T represents the total evaluation time; α i is the physiological response coefficient of crew member i, indicating the degree of influence of physiological monitoring; β i is the attenuation coefficient of crew member i, indicating the attenuation rate of his physiological response over time; B i (t) is the physiological state value of the i-th crew member at time t, ranging from [0,1], indicating the state from rest to fatigue; R i (t) is the workload value of crew member i at time t, with a value range of [0,∞), which is determined by the amount of tasks completed and the intensity of work; γ is the environmental impact coefficient, which indicates the impact of environmental factors on workload; E(t) is the environmental parameter value at time t; M is the number of types of environmental factors; C jis the influence coefficient of the jth environmental factor, and its value range is [0,1]; δ j is the periodic parameter of the jth environmental factor; ∈ is a constant to prevent the denominator from being zero.

[0023] The sensor network dynamically adjusts the sampling frequency through an adaptive algorithm to adapt to monitoring requirements in different working environments.

[0024] Among them, the following formula is used to adjust the frequency: s (t) = f base ·(1+α·ΔE(t)+β·ΔD(t)), where, f s (t) represents the sampling frequency at time t; f base represents the basic sampling frequency, which indicates the sampling frequency under a standard environment; α is the environmental variation coefficient, which indicates the degree of influence of environmental parameter changes on the sampling frequency; ΔE(t) represents the amount of change in environmental parameters at time t; β is the data variation coefficient, which indicates the degree of influence of monitoring data changes on the sampling frequency; ΔD(t) represents the amount of change in monitoring data at time t.

[0025] The method also includes that the central data processing platform is provided with an early warning mechanism, which will automatically issue an early warning notification once it is monitored that the crew's workload exceeds a set safety threshold. The early warning notification includes but is not limited to audio alarms and mobile application push notifications to ensure that the crew can take timely response measures.

[0026] The present invention also proposes a real-time workload monitoring system based on a sensor network, comprising:

[0027] Multiple types of sensors are arranged at corresponding locations within the vessel, including physiological sensors, environmental sensors, and work task sensors;

[0028] A sampling frequency setting module, which is used to set the sampling frequency for each sensor;

[0029] The central data processing platform is used to receive the data collected by the sensors and pre-process the received sensor data;

[0030] The workload judgment module is used to analyze the preprocessed data using the first neural network model of the machine learning algorithm to identify the workload status of the crew.

[0031] Compared with the prior art, the present invention has the following advantages:

[0032] The sensor network of this invention consists of multiple sensor nodes that collect real-time physiological and environmental data from crew members and transmit this data to a central processing platform via wireless communication. This allows for comprehensive monitoring of crew members' work and living conditions, with data transmitted in real time to inform decision-making.

[0033] The combination of cloud computing and stream processing technologies provides powerful technical support for real-time crew monitoring. The high computing and storage capabilities of cloud computing platforms make it possible to process massive amounts of data, while stream processing frameworks (such as Apache Kafka and Apache Flink) enable rapid analysis of real-time data, ensuring its timeliness and accuracy. This architecture enables ship managers to obtain timely information on crew health and work status, thereby optimizing resource allocation and work scheduling.

[0034] Data analysis based on real-time monitoring can provide a scientific basis for crew health management and work arrangements. For example, by analyzing the relationship between physiological responses and workload, managers can develop personalized work plans, arrange rest periods appropriately, and prevent fatigue and health problems. Furthermore, by monitoring environmental factors, managers can promptly adjust navigation strategies to minimize negative impacts on crew members. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The above and other objects, features and advantages of the exemplary embodiments of the present disclosure will become readily understood by reading the following detailed description with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present disclosure are shown in an illustrative and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0036] Figure 1 FIG. 4 is a flow chart illustrating a method for real-time workload monitoring based on a sensor network according to an embodiment of the present invention. DETAILED DESCRIPTION

[0037] To make the objectives, technical solutions, and advantages of the present invention more apparent, the present invention will be further described in detail below with reference to the accompanying drawings. It is apparent that the embodiments described are only some, not all, of the present invention. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without creative effort are intended to fall within the scope of protection of the present invention.

[0038] The terms used in the embodiments of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention. The singular forms "a," "an," "the," and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, and unless the context clearly indicates otherwise, "a plurality" generally includes at least two.

[0039] It should be understood that although the terms "first," "second," "third," etc. may be used to describe "...," these "..." should not be limited to these terms. These terms are merely used to distinguish "...." For example, "first..." could also be referred to as "second...", and similarly, "second..." could also be referred to as "first..." without departing from the scope of the present invention.

[0040] It should be understood that the term "and / or" as used herein is merely a description of the relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " in this document generally indicates that the associated objects are in an "or" relationship.

[0041] As used herein, the words "if" and "if" may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to the determination" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)," depending on the context.

[0042] It should also be noted that the terms "include," "comprises," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a product or device comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such product or device. In the absence of further limitations, an element defined by the phrase "comprises a..." does not exclude the presence of other identical elements in the product or device comprising the element.

[0043] The optional embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0044] Example 1

[0045] like Figure 1 As shown, the present invention discloses a real-time workload monitoring and evaluation method based on a sensor network, comprising the following steps:

[0046] Step S101: arranging multiple types of sensors in a ship, wherein the multiple types of sensors include physiological sensors, environmental sensors, and work task sensors;

[0047] Step S103: For each sensor, set a unified basic sampling frequency;

[0048] Step S105: Transmitting the collected data to a central data processing platform via the sensor network;

[0049] Step S107: pre-processing the received sensor data on the central data processing platform;

[0050] Step S109: Use the first neural network model of the machine learning algorithm to analyze the preprocessed data to identify the workload status of the crew.

[0051] The present invention combines physiological sensors (such as heart rate monitoring), environmental sensors (such as temperature, humidity, and noise), and task sensors (such as task completion time) to conduct comprehensive monitoring. This multi-dimensional data collection can better reflect the crew's working status and environmental impact.

[0052] Big data analytics technology processes sensor data in real time to generate workload assessment reports. Managers can use these reports to make timely decisions, optimize crew work arrangements, and improve overall work efficiency.

[0053] The system sets thresholds and automatically issues warnings if it detects abnormal workloads or unsuitable environmental conditions. This mechanism not only alerts the crew in a timely manner but also notifies managers to take necessary measures to ensure the health and safety of the crew.

[0054] Example 2

[0055] The present invention proposes a real-time workload monitoring method based on a sensor network, comprising the following steps:

[0056] Step S101: arranging multiple types of sensors in a ship, wherein the multiple types of sensors include physiological sensors, environmental sensors, and work task sensors;

[0057] Step S103: For each sensor, set a unified basic sampling frequency;

[0058] Step S105: Transmitting the collected data to a central data processing platform via the sensor network;

[0059] Step S107: pre-processing the received sensor data on the central data processing platform;

[0060] Step S109: Use the first neural network model of the machine learning algorithm to analyze the preprocessed data to identify the workload status of the crew.

[0061] The physiological sensors include a heart rate monitor and a blood oxygen saturation sensor, the environmental sensors include a temperature sensor, a humidity sensor and a noise sensor, and the work task sensors include a task completion time recorder and a work intensity evaluator.

[0062] Wherein, the step S105 includes:

[0063] Each sensor packages the real-time data through the producer mode and sends it to Kafka of the central processing platform;

[0064] The Kafka stores the received real-time data stream in a distributed log.

[0065] Wherein, the step S107 includes:

[0066] The Flink of the central processing platform processes each message sent from the Kafka and performs real-time analysis.

[0067] The preprocessing includes using filtering algorithms to remove noise and ensure the validity of the data.

[0068] The sensor network consists of multiple sensor nodes that monitor workload-related parameters (such as physiological data and environmental data) in real time. Each sensor node sends the collected data to a central data processing platform via a wireless protocol (such as Wi-Fi, Zigbee, or LoRa).

[0069] Data is sent to Apache Kafka through the producer mode. Each sensor node acts as a data producer, packaging real-time data and sending it to a specific Kafka topic for subsequent processing.

[0070] In Kafka, data is organized in topics, supporting high-throughput and low-latency data transmission. Kafka is responsible for receiving real-time data streams from sensors and storing them in a distributed log.

[0071] Kafka provides a data persistence mechanism that ensures that data can be retained and restored even in the event of a system failure. Data can be replayed as needed before being processed to ensure data reliability.

[0072] Apache Flink acts as a Kafka consumer, subscribing to relevant topics to obtain real-time data streams. Flink can process every message sent from Kafka and perform instant analysis.

[0073] Flink provides a flexible data processing pipeline, allowing users to define complex processing logic, including:

[0074] Filter to remove unnecessary data and retain key information.

[0075] Mapping: converting or formatting data.

[0076] Aggregate: perform summary operations on data, such as calculating the average, maximum value, etc.

[0077] Flink supports event time processing and can process data based on timestamps. This enables the system to accurately handle delayed data and perform time window analysis, thereby improving the accuracy of real-time monitoring.

[0078] Data processed by Flink can generate instant analysis results, which users or systems can use to conduct real-time monitoring, such as monitoring workload trends and detecting anomalies.

[0079] Based on real-time analysis results, the system can make dynamic decisions. For example, when it detects an abnormally high workload, the system can automatically trigger an alert or adjust resource allocation to optimize the workload.

[0080] The first neural network model in step S109 is implemented by a multi-layer perceptron structure, specifically including an input layer, a hidden layer and an output layer, wherein the hidden layer adopts a ReLU activation function.

[0081] The first neural network model uses the following formula to evaluate the crew's real-time workload:

[0082] Among them, H i represents the real-time workload evaluation value of crew member i; T represents the total evaluation time; α i is the physiological response coefficient of crew member i, indicating the degree of influence of physiological monitoring; β i is the attenuation coefficient of crew member i, indicating the attenuation rate of his physiological response over time; B i (t) is the physiological state value of the i-th crew member at time t, ranging from [0,1], indicating the state from rest to fatigue; R i (t) is the workload value of crew member i at time t, with a value range of [0,∞), which is determined by the amount of tasks completed and the intensity of work; γ is the environmental impact coefficient, which indicates the impact of environmental factors on workload; E(t) is the environmental parameter value at time t; M is the number of types of environmental factors; C j is the influence coefficient of the jth environmental factor, and its value range is [0,1]; δj is the periodic parameter of the jth environmental factor; ∈ is a constant to prevent the denominator from being zero, usually 0.01.

[0083] Among them, B i (t) is the physiological state function, which represents the physiological response of the crew, such as heart rate, blood oxygen level, etc. i (t) is the workload function, which records the crew's task completion and intensity. E(t) is the environmental parameter function, which covers the impact of temperature, humidity, noise, etc. on workload. i The value range of is [0,∞), which represents the real-time workload assessment value of the i-th crew member. The higher the value, the greater the workload and the potential health risk. It represents a cumulative evaluation of the workload of crew member i within the total time T, which can reflect the changing trend of his workload over time.

[0084] Among them, α i Regression analysis was performed by measuring the participants' physiological responses (such as heart rate, blood pressure, etc.) under different workloads.

[0085] Use historical data to analyze the attenuation coefficient β, which can be expressed by the following formula: Among them, RE t represents the reaction value at time t, and RE0 represents the initial reaction value.

[0086] Experiments were conducted under different environmental conditions, and changes in physiological responses and workload were recorded to obtain the environmental impact coefficient γ, where ΔR represents the change in workload, and ΔE represents the change in environmental factors.

[0087] Determine the influence coefficient C of each environmental factor through historical data analysis j ,in R k is the workload under specific environmental factors j, R base is the benchmark workload and N is the number of samples.

[0088] Use ARIMA model to analyze time series data, identify periodic components, and obtain δ j ,in, Among them E j (t) is the value of the jth environmental factor at time t, and F is the cycle length.

[0089] The sensor network dynamically adjusts the sampling frequency through an adaptive algorithm to adapt to monitoring requirements in different working environments.

[0090] Among them, the following formula is used to adjust the frequency: s (t) = fbase ·(1+α·ΔE(t)+β·ΔD(t)), where, f s (t) represents the sampling frequency at time t; f base represents the basic sampling frequency, which indicates the sampling frequency under a standard environment; α is the environmental variation coefficient, which indicates the degree of influence of changes in environmental parameters on the sampling frequency; ΔE(t) represents the change in environmental parameters at time t (such as temperature, humidity, noise, etc.); β is the data variation coefficient, which indicates the degree of influence of changes in monitoring data on the sampling frequency; ΔD(t) represents the change in monitoring data at time t (such as physiological parameters, workload, etc.).

[0091] Environmental parameters include temperature, humidity, and noise. ΔE(t) can be defined as the change in these parameters at time t. Assuming these parameters are recorded at consecutive time points t and t-1, they can be calculated using the following formula:

[0092]

[0093] ;E temp (t) represents the temperature value at time t; E hum (t) represents the humidity value at time t; E noise (t) represents the noise value at time t.

[0094] The monitoring data include physiological parameters and workload, etc. Similar to the change in environmental parameters, ΔD(t) can be defined as the change in these monitoring data at time t: Among them, D phys (t) represents the physiological parameter value at time t (such as heart rate, blood oxygen, etc.); D load (t) Workload value at time t (such as task intensity, completion amount, etc.).

[0095] The environmental change coefficient α can be obtained through historical data regression analysis. The specific steps are as follows:

[0096] Collect environmental data over a period of time and the corresponding sampling frequency.

[0097] Use linear regression or other statistical methods to analyze the impact of environmental changes on sampling frequency and obtain the regression coefficient.

[0098] The method for obtaining the data variation coefficient β is similar to that for α:

[0099] Collect historical monitoring data and corresponding sampling frequencies.

[0100] Linear regression analysis was performed to establish the relationship between changes in monitoring data and sampling frequency.

[0101] The specific adjustment methods are as follows:

[0102] Adjustment method

[0103] Real-time monitoring uses sensors to continuously collect environmental and monitoring data.

[0104] Dynamic calculation, at each time point t, calculate the new sampling frequency f based on the current ΔE(t) and ΔD(t) s (t).

[0105] Frequency setting, based on the calculated f s (t), adjust the sampling configuration of the sensor. If f s As (t) increases, the sensor collects data at a higher frequency; conversely, the sampling frequency is reduced to save energy and storage. Through this adaptive algorithm, the sensor network can dynamically adjust the sampling frequency based on varying environmental conditions and monitoring requirements. This approach not only ensures real-time data accuracy but also improves the system's energy efficiency and resource utilization, providing strong support for crew health monitoring.

[0106] The method also includes that the central data processing platform is equipped with an early warning mechanism. Once the workload of the crew is detected to exceed the set safety threshold, an early warning notification will be automatically issued. The early warning notification includes but is not limited to audio alarms and mobile application push notifications to ensure that the crew can take timely countermeasures. The system sets a threshold. Once an abnormal workload (such as excessive heart rate, long working hours, etc.) or unsuitable environmental conditions (such as excessive temperature, excessive noise) is detected, the system will automatically issue an early warning. The early warning information will be promptly conveyed to the crew and managers through mobile applications or other communication methods, prompting them to take necessary measures to ensure the safety and health of the crew.

[0107] The method also includes long-term tracking of the health status of crew members, establishing a health database, combining machine learning technology to predict the health risks of crew members, and providing decision support for managers.

[0108] Among them, multi-dimensional visualization of data is achieved through multiple user interfaces. Crew members and managers can view real-time monitoring data and historical records through mobile applications, which facilitates health management and workload adjustment.

[0109] Example 3:

[0110] The present invention also proposes a real-time workload monitoring system based on a sensor network, comprising:

[0111] Multiple types of sensors are arranged at corresponding locations within the vessel, including physiological sensors, environmental sensors, and work task sensors;

[0112] A sampling frequency setting module, which is used to set the sampling frequency for each sensor;

[0113] The central data processing platform is used to receive the data collected by the sensors and pre-process the received sensor data;

[0114] The workload judgment module is used to analyze the preprocessed data using the first neural network model of the machine learning algorithm to identify the workload status of the crew.

[0115] Example

[0116] An embodiment of the present disclosure provides a non-volatile computer storage medium, wherein the computer storage medium stores computer-executable instructions, and the computer-executable instructions can execute the method steps described in the above embodiment.

[0117] It should be noted that the computer-readable medium mentioned above in the present disclosure may be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or component, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, device, or component. In the present disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to wires, optical cables, RF (radio frequency), etc., or any suitable combination thereof.

[0118] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0119] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0120] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the module, program segment, or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of the boxes in the block diagram and / or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0121] The units involved in the embodiments described in this disclosure may be implemented in software or hardware, wherein the name of a unit does not necessarily limit the unit itself.

[0122] The above introduces the preferred embodiments of the present invention, which is intended to make the spirit of the present invention clearer and easier to understand, and is not intended to limit the present invention. Any modifications, replacements, and improvements made within the spirit and principles of the present invention should be included in the scope of protection outlined by the claims attached to the present invention.

Claims

1. A real-time workload monitoring method based on a sensor network, characterized in that: The following steps are involved: Step S101: arranging multiple types of sensors in a ship, wherein the multiple types of sensors include physiological sensors, environmental sensors, and work task sensors; Step S103: For each sensor, set a unified basic sampling frequency; Step S105: Transmitting the collected data to a central data processing platform via the sensor network; Step S107: pre-processing the received sensor data on the central data processing platform; Step S109: Use the first neural network model of the machine learning algorithm to analyze the preprocessed data to identify the workload status of the crew.

2. The method according to claim 1, wherein The physiological sensors include a heart rate monitor and a blood oxygen saturation sensor, the environmental sensors include a temperature sensor, a humidity sensor and a noise sensor, and the work task sensors include a task completion time recorder and a work intensity evaluator.

3. The method according to claim 1, wherein: The step S105 includes: Each sensor packages the real-time data through the producer mode and sends it to Kafka of the central processing platform; The Kafka stores the received real-time data stream in a distributed log.

4. The method according to claim 3, wherein: The step S107 includes: The Flink of the central processing platform processes each message sent from the Kafka and performs real-time analysis.

5. The method according to claim 1, wherein: The first neural network model in step S109 is implemented by a multi-layer perceptron structure, specifically including an input layer, a hidden layer and an output layer, wherein the hidden layer adopts a ReLU activation function.

6. The method according to claim 5, wherein: The first neural network model uses the following formula to evaluate the crew's real-time workload: Among them, H i represents the real-time workload evaluation value of crew member i; T represents the total evaluation time; α i is the physiological response coefficient of crew member i, indicating the degree of influence of physiological monitoring; β i is the attenuation coefficient of crew member i, indicating the attenuation rate of his physiological response over time; B i (t) is the physiological state value of crew member i at time t, ranging from [0, 1], indicating the state from rest to fatigue; R i (t) is the workload value of crew member i at time t, with a value range of [0, ∞), which is determined by the amount of tasks completed and the intensity of work; γ is the environmental impact coefficient, which represents the impact of environmental factors on workload; E(t) is the environmental parameter value at time t; M is the number of types of environmental factors; C j is the influence coefficient of the jth environmental factor, and its value range is [0, 1]; δ j is the periodic parameter of the jth environmental factor; ∈ is a constant to prevent the denominator from being zero.

7. The method according to claim 1, wherein: The sensor network dynamically adjusts the sampling frequency through an adaptive algorithm to adapt to monitoring requirements in different working environments.

8. The method according to claim 7, wherein: Use the following formula to adjust the frequency: s (t) = f base ·(1+α·ΔE(t)+β·ΔD(t)), where, f s (t) represents the sampling frequency at time t; f base represents the basic sampling frequency, which indicates the sampling frequency under a standard environment; α is the environmental variation coefficient, which indicates the degree of influence of environmental parameter changes on the sampling frequency; ΔE(t) represents the amount of change in environmental parameters at time t; β is the data variation coefficient, which indicates the degree of influence of monitoring data changes on the sampling frequency; ΔD(t) represents the amount of change in monitoring data at time t.

9. The method according to claim 1, wherein: The method also includes that the central data processing platform is provided with an early warning mechanism, which will automatically issue an early warning notification once it is monitored that the crew's workload exceeds a set safety threshold. The early warning notification includes but is not limited to audio alarms and mobile application push notifications to ensure that the crew can take timely response measures.

10. A real-time workload monitoring system based on a sensor network, comprising Multiple types of sensors are arranged at corresponding locations within the vessel, including physiological sensors, environmental sensors, and work task sensors; A sampling frequency setting module, which is used to set the sampling frequency for each sensor; The central data processing platform is used to receive the data collected by the sensors and pre-process the received sensor data; The workload judgment module is used to analyze the preprocessed data using the first neural network model of the machine learning algorithm to identify the workload status of the crew.

Images