Research and Application of Port Crane Health Assurance System

Considering portal crane is subject to long service life, harsh working conditions, high damage rate and regular maintenance, to ensure its healthy and reliable operation, an on-line remote monitoring and preventive maintenance and health security system is established based on technologies such as Internet of Things, cloud service platform and industrial big data analysis. The health security system can tell whether the crane is healthy or not with the collected data and give suggestions on preventive maintenance. Moreover, combined with the application network and overall framework of health security system , some functional modules are designed for full lifecycle of the crane, including data acquisition and integration, health analysis optimization based on big data, collaborative application among enterprises based on public Internet, cloud service system and security system. The paper further discusses key technologies of health security system such as remote monitoring and automat ic identification system for cracks in in crane’s steel structure, health security system based on cloud service platform, and crane fault diagnosis and prediction analysis optimization system based on deep learning and big data analysis. The paper ends with presenting problems and challenges in the implementation of health security system based on the research progress of health security system for portalcrane.
The automated container terminal caters to the development needs of industrialization and informatization, and China has successfully completed the Xiamen Yuanhai fully automated terminal [2]. The traditional operation method of relying on the driver to manually operate the crane in the crane cab has gradually changed to the operation in the semi-automatic assistance mode by the personnel in the remote control room. Due to the separation of man and machine, it is no longer possible to rely on the original driver’s direct perception (hearing, touch, vision, etc.) to detect crane failures during operation.
Port lifting equipment has a complex mechanical structure and involves many technical fields. Various damage mechanisms, long service periods, harsh working conditions, high damage rates, and need for regular maintenance have led to increased maintenance difficulties, increased maintenance tasks, and professional skills. Maintenance personnel. The user needs to perform periodic manual inspections on key locations in accordance with the equipment maintenance manual to ensure that the structure and equipment are safe and reliable. However, there will be phenomena such as failure to follow the maintenance manual for regular inspections, missed inspections of some key locations, and low professionalism of the inspectors. Ensure the accuracy of each test. Especially when the equipment has a problem that cannot be solved by the port maintenance personnel during the operation of the crane, the user can only ask for help by email or phone. This unintuitive method and regional differences in language and culture often lead to low efficiency in solving problems. After the fault is finally confirmed, a maintenance engineer is sent to the customer site for inspection and repair, making the entire maintenance process cycle very long [3].
At present, the port cranes in service still lack a complete and unified health protection information management platform. Enterprises cannot obtain the operation information of sold products. It is difficult to guarantee the timeliness, initiative, on-site and foresight of service work. The current status of after-sales service is basically It is still the traditional model of headache medicine. In addition, research on online remote monitoring and preventive maintenance of port cranes based on technologies such as the Internet of Things technology, cloud service platform and industrial big data analysis has just started outside China. How to effectively monitor the operation of cranes in real time It is of great practical significance to carry out preventive maintenance services, reduce downtime losses, increase the added value of products, and strengthen the bond of interest with customers.

1. The overall composition of the health protection system

The health protection system is based on the industrial Internet technology, which integrates existing crane equipment with intelligent equipment, intelligent systems and intelligent decision-making, and provides customers with a complete set of port crane operation, management and maintenance solutions, which can promote port crane equipment and Networked collaboration and service-oriented extension of related industries.
The network structure design of the health protection system covers various crane equipment, terminal control centers, production bases and corporate data centers in the port, and runs through the life cycle of cranes related to production, supporting, construction, operation and maintenance and maintenance.
At the local equipment level of port cranes, it is necessary to add dedicated data communication acquisition equipment and video monitoring systems to the traditional control system, effectively integrate wired and wireless communication technologies, and strengthen the interconnection capabilities between monitoring equipment, sensors, control systems and management systems. The crane local equipment layer uses the existing CMS (Crane Monitoring System) system to obtain the sensor data filtered by the collection system to complete the local simple fault diagnosis function.
In addition to providing the basic functions of RCMS, the terminal control center also needs to add a data collection server. The data collection server is deployed in the port server room for data collection of sensors, working conditions, operation and maintenance, and uploading data to Cloud. The data collection server completes the online fault diagnosis and prediction function according to the mechanical or structural finite element model, historical data matching, locally deployed machine learning and migrated big data computing library, and returns fault and failure information to the local crane layer when necessary .
The terminal can be connected to the production base and enterprise data center through a dedicated VPN network or data cloud. Through data cloud technology, it is convenient for terminals and enterprises to carry out effective integration and analysis of massive industrial data based on cloud computing and big data technology, and promote the intelligent decision-making of port crane health detection. The enterprise data center uses OPC service agreement to subscribe to the operating data of the terminal and production base through dedicated lines and data cloud networks to complete the data collection function of real-time monitoring, thereby achieving the crane health protection function.
As a result, a big data cloud technology platform based on front-end data collection, intermediate high-speed transmission, back-end analysis and evaluation, and feedback results is established. The use of cloud-based services to realize the port crane health protection system, and providing users with on-demand and easily scalable services through the Internet will effectively solve various problems faced by traditional services. Users can reduce the cost of investment in basic equipment, and reduce system management, IT expenditures. By collecting and analyzing user crane data, you can obtain product quality, spare parts and other information. On the one hand, you can continuously improve product quality based on this, and on the other hand, provide users with real-time and forward-looking services to establish a technical foundation.
The port crane health protection system adopts physical network technology and is functionally divided into an information perception layer, an information transmission layer, an information service layer, and an information value-added layer.
The information perception layer can solve the problem of data acquisition and implement processes from equipment, control, and sensing. Among them, the crane equipment of the port includes such as shore container cranes, tire cranes, rail cranes, rotating mechanisms, reducers, etc.; the control layer completes the drivers, inverters, relays, and industrial computers involved in the on-board local control, and completes the on-board local monitoring The acquisition system; the sensor layer to obtain sensory information such as temperature, strain, acceleration, vibration, and position.
The information transmission layer is to solve the problem of data transmission. It uses various industrial buses, video branching, Ethernet, communication gateways, switches, 3G/4G, WIFI and other technologies to build equipment, control systems, acquisition systems, and remote centralized control centers. Big data cloud, data interconnection network between enterprises.
The information service layer is the business processing based on the filtering and preprocessing of the perception data, including the basic functions of the data layer for data storage, visualization, alarm, log, historical playback, parameter model, data analysis, etc., and the service layer, which provides online monitoring , Failure analysis, failure analysis, failure prediction, life prediction, maintenance recommendations, monitoring evaluation and other services. The information service layer implements basic and basic services for health protection, reflecting the interrelationship between collection and diagnosis, diagnosis and business, and business and control.
The information value-added layer is the improvement and sublimation of the information service layer, providing functions such as network optimization, fleet optimization, performance optimization, asset allocation, operation and maintenance guarantee, and system guarantee. Under the technical support of the Internet, big data, cloud technology, and Internet of Things technology, the information value-added layer integrates the crane equipment health protection system with the enterprise’s ERP (Enterprise Resource Planning) system to facilitate the equipment enterprise and enterprise between the terminal and the enterprise. Establish business interconnection relationships between industries, effectively connect enterprises and customers in real time, connect people in various workplaces, and provide smarter, high-quality services and security.

2. Main functions of the system

2.1 Data collection and integration for the full life cycle of port cranes

The management of the whole life cycle process of port cranes requires the integration of information and processes in the whole life cycle from crane demand, planning, design, production, transportation, debugging, operation, use, maintenance, until recycling and disposal [4]. Stage business and data information, establish a unified crane data model, and realize the informationized data management process for the life of the crane.
In order to implement the crane health protection system, the installation, wiring, construction, debugging, verification and replacement of the sensor layer of the newly built port crane need to be considered from the design stage. Other modified crane equipment may be considered for ease of installation, installation points and easy Factors such as the point of failure. The monitoring of key points of the crane needs to consider the sensitivity, strain capacity and dynamic response range of the measuring points, combined with the model to realize the analysis of the physical characteristics of the key position stress, strain and structural vibration before selecting. In order to realize the economy, practicability and accuracy of the health protection system, the selection points of the sensors need to be increased or decreased according to the user’s needs, so that differentiated, global and diversified solutions can be formed.
In the manufacturing stage, the crane accessories and construction data need to be manually entered, including the origin, model, warehousing time, installation time, installation, welding, painting and other related information throughout the construction process; base debugging and transportation stage, data collection work It is completed by crane-specific data acquisition equipment, and the data is cached in the local airborne industrial computer. The shipping stage of a large crane can range from a few days to several months. At this stage, some sensors need to be turned on to record the impact of wind and wave loads on the steel structure during the transportation process, which can more accurately serve the later health diagnosis; in the service stage, Crane-specific data collection equipment and port data collection and transmission equipment complete the complete data collection work. A data acquisition system oriented to the full life cycle should have the following functions:
1) The special data collection equipment for port cranes can meet the requirements of various local data collection of cranes. It should also have data preprocessing capabilities, and be able to perform some statistics and data screening work to form statistical data or statistical database records. Each type of crane can complete a variety of different statistical needs according to the characteristics of the project, including statistics of important institutions, equipment running time, running times, working cycle statistics, equipment failure time statistics, maintenance time statistics, etc.
2) The special collection equipment for port crane data has certain data cache requirements and power-down protection functions. In the service phase, when there is a problem in the communication between it and the port data storage device, it can cache real-time data. Various statistical data are also required to be stored locally for a period of time until they are read by the port data storage device.
3) Each port crane dedicated data collection equipment, as well as port data collection and transmission equipment, has a unique identification number and a unified time service, so that different crane equipment of different docks can be integrated in the data processing center.
4) Special data collection equipment for port cranes, as well as port data collection and transmission equipment, need to send real-time data and other statistical data in accordance with unified data communication requirements. The crane-specific data collection equipment sends data directly through the 3G/4G network, or the data can be aggregated to the port and then sent uniformly.

2.2 Health analysis optimization based on big data

After the port crane health assessment system collects the data of the crane’s life cycle, it provides a health report through the online fault diagnosis program and the calculation of the big data cloud platform. Based on the health diagnosis report, users can more fully grasp the crane’s use status, guide the daily maintenance of the crane, and record the daily maintenance of the structure Status, it is convenient for the terminal to carry out integrated information management. The establishment of a complete crane health protection system will greatly maintain the integrity of crane equipment, minimize the incidence of equipment failures, and maximize the economic benefits of manufacturers and terminal users.
At the wharf, sudden unforeseen failures of cranes include structural cracks and damage to major components such as reducers and motors. This type of failure will cause the equipment to shut down due to the long production time of spare parts. Preventive maintenance is the trend of crane equipment health diagnosis. The crane health protection system includes online health estimation and offline diagnosis methods based on big data analysis. Big data health analysis is a supplement and improvement to traditional online health prediction. It combines data mining, machine learning, and deep learning algorithms to predict equipment failures, and the accuracy of prediction can be effectively improved.
Online crane health prediction module, based on the crane load capacity, trolley position and other stress sensor data and CMS historical data, through the analysis of the recorded load spectrum, the stress spectrum of the steel structure of the crane is obtained. The diagnostic software calculates the theoretical life of the key locations of the crane steel structure, and quickly and accurately evaluates the usage and real-time health status of the crane steel structure. The mechanism detection of the crane relies on the vibration signal of the equipment. By collecting the abundant vibration signals of the crane rotating equipment such as bearings, gears, reducers, etc., the signal characteristic information is extracted to identify the fault type of the equipment. The online health estimation module is integrated in the CMS system to provide an early warning function, which can issue an early warning to users, remind users to pay special attention to the location of abnormal points, increase the frequency of inspections, and remind users to prepare for replacement of spare parts. For critical locations where the actual life expectancy exceeds the theoretical life expectancy, a danger signal will be sent to the user immediately, reminding the user to prepare spare parts as soon as possible, and stop production for maintenance and replacement at the appropriate time.
The offline health diagnosis method of big data analysis extends the traditional fault diagnosis method based on model, modal, fault tree, and expert system [5]. Based on massive port crane production, sensors, working conditions, accessories and other data, combined with daily maintenance experience judgments, targeted parts fatigue life tests, preventive measures for spare parts, daily inspections and routine maintenance, history Carry out data modeling and big data analysis for failures and failure-related information of products of the same type, use the analysis results to form preventive maintenance decisions, and improve the health prediction function of port cranes. The big data offline analysis health diagnosis system gives a diagnosis report according to the port production scheduling plan, which can subvert the traditional simple way of only relying on the maintenance manual for maintenance, and accurately plan the maintenance cycle and repair parts.

2.3 Collaborative applications between enterprises based on the public Internet

The collaborative application between enterprises based on the public Internet is deployed on the basis of IPv6 and high-performance network technology to form a set of remote monitoring and management system for port crane equipment on the enterprise data center server. By remotely monitoring the real-time situation of port cranes, using the health protection system to predict upcoming failures, the coordination between manufacturing companies and ports can be resolved. Using cloud platform technology to realize online monitoring of the monitoring system, realize the interconnection and intercommunication between equipment and personnel, build a safe, stable and fast monitoring system, effectively collect the main status information of the equipment, integrate multiple heterogeneous data sources, and establish a A large-scale online monitoring platform effectively establishes collaboration between enterprises and port enterprises.
Crane remote online monitoring is a crane equipment management system deployed in the cloud. The remote online monitoring of cranes has different portals for the company’s internal personnel and port users. For company users, it can be accessed through the global crane online monitoring portal, supplemented by related value-added services. For port users, they are accessed through a separate port online monitoring portal, and the data is concentrated on environmental data in the integrated system, such as the status of the organization, the status of spreaders, and the monitoring of third-party systems.
The collaborative platform can realize the macro status data collected by each crane by region according to the geographical location and map, including the crane status signal and the status signal of each monitoring subsystem, such as the crane CMS status signal, the mechanical status monitoring subsystem (refers to the mechanical structure, The organization’s dedicated online monitoring local system) operating status signal, tire pressure monitoring subsystem operating status signal, speed monitoring subsystem, etc. When the communication of each device fails, it will be automatically recorded in the database and the user will be notified in time. In addition, the port online monitoring system also includes various operating statistical reports, such as failure statistical reports, working cycle statistical reports, operating time statistical reports, etc. It also has a backup function for crane data collection equipment and port data storage equipment. Provide system and data recovery functions.

2.4 Cloud service system security assurance system

The construction of the safety guarantee system is mainly aimed at the safety guarantee system of the port crane health guarantee, industrial Internet networking and integrated application, and coordination and management of software and hardware, as well as the generated data and information. In order to ensure the security of the cloud service system, prevent important data leakage and network attacks, establish a robust security mechanism to ensure the security of monitoring data in the collection process, network transmission, and data storage, and reduce the cloud service system to the greatest extent. Security threats, improve the continuity of cloud services, and ensure the healthy and sustainable development of cloud service applications.
The health protection system requires the terminal to integrate enterprise-level networks, cloud platforms, and industrial control system networks, and realize the connection between industrial control systems and enterprise information management systems, cloud service platforms, big data analysis, and big data monitoring systems through the public Internet. Data sharing to realize the automated operation of industrial facilities, process monitoring and control, and business process management and control. However, in the interconnection of enterprise-level information networks, cloud platforms, and industrial control networks, there are often hidden security hazards or risks. If common security threats (such as viruses, worms, Trojan horses, etc.) in the public network are introduced into the industrial control network , The industrial network will be attacked by viruses, unauthorized access, information leakage, etc., which will seriously affect the normal operation of the terminal.
With the expansion of software scale, terminal application systems such as SCADA, PLC, IED, RTU, and real-time I/O need to be designed with network robustness and information security issues in mind. In the terminal network, it is necessary to standardize the electrical room and the centralized control center for potential multi-point intrusion infection sources or hidden attacks, such as mobile portable devices, USB devices, serial connections, wireless connection systems, and unauthorized devices Access and eavesdropping, etc. In addition, necessary network isolation and segmentation mechanisms are required in the industrial control network, cloud and enterprise network. If a one-way transmission gateway is appropriately added, any vulnerable area in the system will be compromised, and the risk will not spread to other areas. .
Cloud crane monitoring data needs to adopt the idea of ​​combining symmetric encryption and asymmetric encryption to solve the problem of data security storage in cloud computing. The main research of monitoring data transmission security includes the reliability of data transmission in the cloud computing platform, such as the use of TCP/IP protocol multi-point parallel transmission, cutting large data blocks, and then using parallel transmission to reduce network delay [6, 7].

3. Research on key technologies

3.1 Remote monitoring and automatic identification system for steel structure cracks of port cranes

Insufficient local strength and rigidity of the crane steel structure will cause cracks. The crack detection of port cranes can use the resistance-based resistance strain method, or use fiber grating strain sensing to measure the strain force on the surface of the rigid structure. The analysis method uses the method of crack propagation resistance to estimate crack. Due to the relatively large size of port cranes, the number of sensor points cannot cover the entire machine, and it is difficult to evaluate the structural life of the entire machine.
The remote monitoring and automatic identification system for crane steel structure cracks mainly uses image recognition technology to automatically identify steel structures. By installing cameras at key points of crane crack monitoring, real-time online monitoring of key points of crane cracks. The system can analyze steel structure crack indicators, transmit analysis results in real time, replace worker inspections, and can remotely query the real-time structure of steel structures to improve workers’ work efficiency.
The key to the design of the automatic recognition system is to ensure that the image recognition camera can capture complete crack information. This not only requires a reasonable installation position of the image recognition system camera, but also a clear understanding of the basic knowledge of cracks. Since the sample information of cracks in the steel structure is relatively small, it is necessary to continuously accumulate crack pictures, and gradually provide the accuracy of crack recognition through model matching or deep network. The automatic identification system takes the automatic identification of cracks as the starting point. The friendly interactive interface can help remote personnel to confirm the on-site situation through the video screen even when automatic confirmation is not possible, so as to achieve complete identification, thereby greatly reducing the work intensity of the operator.

3.2 Research on health insurance system based on cloud service platform

The health protection system uses cloud service technology, is designed in accordance with the architecture concept of platform + tools, and covers the overall cloud computing solution of end, management, and cloud, to provide all-round protection for the crane health protection system.
IaaS (Infrastructure as a Service) service is at the bottom of cloud computing and can directly provide highly elastic available services for upper-layer applications. The remote monitoring of cranes is based on the IoT technology. Data collection is carried out by arranging sensor devices on the device side, and then encrypted and transmitted to the cloud data center via the network to realize cloud storage and make preliminary preparations for further real-time data analysis. Among them, combining the relevant parameters and real-time characteristics of the crane equipment and its operating environment, designing an efficient and dedicated distributed storage framework and a highly scalable computing model is the key to big data elastic computing and storage services based on IaaS and IoT, and the advantages and disadvantages of the solutions are directly Affect the timeliness, accuracy and availability of remote monitoring. PaaS (Platform as a Service) is in the middle layer of cloud computing, mainly for software development and maintenance, and its scope of influence covers the entire life cycle of the software. The operation and management of crane equipment are shared by multiple different software systems or subsystems, such as crane status signal, machinery status monitoring subsystem, tire pressure monitoring subsystem, speed monitoring subsystem, RCMS crane remote management system, etc. , Due to differences in providers or service aspects, they are all isolated islands of information that operate independently, causing on-site monitoring and fault warning to rely more on personal experience rather than the results of quantitative analysis.
The cloud application SaaS (Software as a Service) provides professional services directly to the specific needs of customers. Since there are two types of crane remote online monitoring users: company internal personnel and port users, the main research content for SaaS cloud services and in-depth exploration of customer value-added services is how to establish a dedicated portal system for different users and customize differentiated services. For company users, it can be accessed through the global crane online monitoring portal. The direct data comes from the integrated system in PaaS, supplemented by related value-added services. For port users, they are accessed through a separate port online monitoring portal, and the data is concentrated on environmental data in the integrated system, such as the status of the organization, the status of spreaders, and the monitoring of third-party systems.

3.3 Crane fault diagnosis and prediction analysis optimization based on deep learning and big data analysis

Port lifting equipment is developing towards complexity and large-scale planning, and the health protection system collects massive amounts of data to reflect the health of the machinery. These machinery-related big data have the characteristics of large capacity, diversity and high speed. Research and use advanced theories and methods to mine information from the big data of machinery and equipment, efficiently and accurately identify the health status of the equipment, and become its health monitoring New problems facing the field.
Traditionally, for example, through the condition monitoring of important parts of cranes such as transmission, encoder, brake, diesel engine, wire rope, etc., the probability model of each parameter change and component damage is described, and the probability model is compared with the current multi-parameter probabilistic state space. , To achieve the realization of the current health status and trend analysis of judging components.

4. Existing problems and challenges

The research on the port crane health protection system is still in its infancy. Traditional mechanical health protection methods cannot be fully applied to the port crane industry. Research on smart sensing, smart detection, smart identification, business coordination, etc. is not sufficient. At present, port cranes The following problems are common in the health protection system:
1) Reasonable placement, layout, quantity, and sensor type of sensors, lack of theoretical guidance, and less experience in actual successful engineering projects;
2) The cost and accuracy of the sensor are difficult to weigh. In order to achieve an effective diagnosis effect, the cost of the system equipment is often too high, and the entire system lacks economic benefits;
3) The layout of sensor lines for port cranes is troublesome, and wireless sensors cannot be well applied in the crane field due to issues such as power supply and communication bandwidth;
4) The service period of port cranes is long, and the application environment is relatively harsh. The sensor needs to be subjected to the test of high temperature, high humidity, high salinity and other environments for a long time outdoors. The sensor is prone to zero drift, loss of accuracy, or even damage.
If the sensor is inaccurate and the collection system does not have a better detection method, the collected bad data will affect the fault diagnosis; the integration of deep learning, cloud computing and other technologies with the port crane industry will lead to the advancement and improvement of crane health protection technology. Depends on the following developments:
1) Continuous research on the mechanism of structural cracks and mechanism damage of port cranes, combined with the continuous development of structural mechanics, fracture mechanics, damage mechanics and other disciplines, to deepen the theoretical understanding of the causes of failures;
2) Strengthen the research of smart sensors, continuously improve the accuracy and intelligence of the sensors, and develop cost-effective sensors with independent calibration, decision-making, and preprocessing to meet the needs of the health protection system for signal collection [17];
3) Research on the algorithms of sensor collection and signal processing, strengthen the collection of crane status and operation and maintenance data, establish data mining and big data analysis models based on port cranes, and carry out research with the help of big data platform tools.

5. Concluding remarks

The service life of port cranes is generally more than 20 years. Under the situation that the market is increasingly saturated and new machine orders are declining, how to provide value-added after-sales service will be a key measure to ensure the sustainable and healthy development of enterprises. At the same time, the life-span management of products must require companies to fully grasp various information about sold products in a timely manner. The port crane health guarantee system will conform to the technological development trend of the port and provide strong technical support for the maintenance and smooth operation of the port’s equipment.
Based on the Internet of Things, cloud service platform and industrial big data analysis to establish a health protection system, the crane will be

---

Related Items