Cross-network-segment centralized monitoring system, method and device for discrete control system

Abstract

The invention relates to a cross-network-segment centralized monitoring system, method and device for a discrete control system, and belongs to the technical field of industrial automation control. Comprising an acquisition module used for acquiring an operation state data packet of a condensate water pump station; the protocol conversion module is used for judging whether the first communication protocol of the acquisition module is consistent with a preset second communication protocol or not; if not, converting the running state data packet into second communication protocol format data; otherwise, conversion is not carried out; the pool liquid level control module is used for outputting a control signal for controlling a water pump to start and stop according to the intermediate data; and the display module is used for displaying the parameters in the running state data packet and the execution process of the control signal in real time. According to the invention, the problems of difficulty in centralized monitoring caused by dispersion of the existing condensate water pump station system and environmental protection and potential safety hazards caused by fault response lag are solved, centralized monitoring is realized, and the management efficiency and the response speed are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial automation control, and in particular to a system, method and device for centralized monitoring of discrete control systems across network segments. Background Art

[0002] The operation of the condensate pumping stations at the gas protection stations faced numerous technical challenges that needed to be addressed. First, the condensate pumping stations were widely distributed, and their IP addresses were not on the same network segment. This made centralized monitoring difficult and limited the efficiency of unified management and scheduling of the entire system. Second, failures in the level gauges and condensate pumps often went undetected, which could easily lead to serious overflows of the gas condensate tanks, resulting in a series of environmental and safety hazards, posing a potential threat to both the company's normal operations and the surrounding environment. Furthermore, the existing control system primarily utilized traditional relay control. This outdated control model no longer met the high efficiency, high precision, and high reliability requirements of modern automation control, severely restricting overall system performance. Furthermore, the decentralized control system not only increased management complexity and costs but also led to inefficient inspection and repair work by on-site maintenance personnel, further impacting equipment operation and causing significant inconvenience to production safety. Finally, the characteristics of the system itself also mean that when the liquid level reaches the alarm condition and the water pump needs to be started, due to the limitations of equipment performance, some condensate may overflow, causing hidden dangers such as environmental pollution. This also reflects to a certain extent the shortcomings of the current system in responding to emergencies. Summary of the Invention

[0003] To this end, the technical problem to be solved by the present invention is to overcome the difficulties in centralized monitoring caused by the dispersion of the condensate pump station system in the existing technology, the environmental and safety risks caused by the delayed fault response, the inability to meet modern automation control requirements, the low management efficiency and the delayed control response.

[0004] In a first aspect, to solve the above technical problems, the present invention provides a cross-segment centralized monitoring system for a discrete control system, comprising:

[0005] The acquisition module is used to obtain the operating status data packet of the condensate pump station;

[0006] a protocol conversion module, configured to determine whether the first communication protocol of the acquisition module is consistent with a preset second communication protocol; if not, converting the running status data packet into data in the second communication protocol format and obtaining intermediate data; otherwise, obtaining the intermediate data directly without performing the conversion;

[0007] A water tank liquid level control module, configured to output a control signal for starting and stopping the water pump according to the intermediate data;

[0008] The display module is used to display the parameters in the operation status data packet and the execution process of the control signal in real time.

[0009] In one embodiment of the present invention, the water tank liquid level control module includes:

[0010] A recording submodule, for collecting and storing the intermediate data and generating historical data according to preset rules;

[0011] A data preprocessing submodule, configured to clean, normalize, and time-series-align the historical data to obtain first data;

[0012] a feature extraction submodule for extracting features from the first data to obtain liquid level variation regularity features and water pump start-stop regularity features; and performing fusion deep feature mining and data correlation analysis based on the liquid level variation regularity features and water pump start-stop regularity features to obtain a feature set;

[0013] A prediction submodule is used to construct a liquid level prediction model and train the liquid level prediction model according to the feature set to obtain a liquid level prediction result;

[0014] The output submodule is used to output a control signal for controlling the start and stop of the water pump according to the liquid level prediction result.

[0015] In one embodiment of the present invention, the process of integrating deep feature mining and data association analysis in the feature extraction submodule includes establishing association rules between liquid level data and related data; and calculating the support and confidence of the association rules.

[0016] In one embodiment of the present invention, the water tank liquid level control module further includes a feedback submodule for monitoring the actual change of the water tank liquid level in real time, evaluating the deviation degree of the current liquid level control, and then dynamically adjusting the prediction model.

[0017] In one embodiment of the present invention, the protocol conversion module includes:

[0018] A data receiving submodule, configured to receive the running status data packet;

[0019] The judgment submodule is used to obtain the first identifier of the first communication protocol and confirm the second identifier of the preset second communication protocol; use a matching function to determine whether the first identifier and the second identifier are the same; if they are the same, directly send the operating status data packet to the switch; otherwise, send a conversion signal.

[0020] In one embodiment of the present invention, the protocol conversion module further includes:

[0021] A data parsing submodule, configured to receive the conversion signal and extract valid information from the operation status data packet;

[0022] A data conversion submodule converts the valid information into data in the format of the second communication protocol according to the specification of the preset second communication protocol, and obtains intermediate data;

[0023] The data sending submodule is used to send the intermediate data to the switch.

[0024] In one embodiment of the present invention, a switch is further included for establishing communication between the water pool level control module, the display module and the protocol conversion module.

[0025] In one embodiment of the present invention, an alarm module is further included for monitoring parameters of the running status data packet in real time, comparing the parameters with the alarm threshold, determining whether an abnormality exists, and issuing an alarm signal in a timely manner when an abnormality is detected.

[0026] In a second aspect, to solve the above technical problems, the present invention provides a method for centralized monitoring of a discrete control system across network segments, comprising:

[0027] Obtain the operating status data packet of the condensate pump station;

[0028] determining whether the first communication protocol for transmitting the operating status data packet is consistent with a preset second communication protocol; if not, converting the operating status data packet into data in the second communication protocol format to obtain intermediate data; otherwise, directly obtaining the intermediate data without performing the conversion;

[0029] Outputting a control signal for starting and stopping the water pump according to the intermediate data;

[0030] The parameter changes in the operating status data packet and the execution process of the control signal are monitored in real time.

[0031] On the third aspect, in order to solve the above technical problems, the present invention provides a cross-segment centralized monitoring device for a discrete control system, including the above-mentioned cross-segment centralized monitoring system for a discrete control system.

[0032] The above technical solution of the present invention has the following beneficial effects compared with the prior art:

[0033] (1) The present invention describes a system, method, and device for centralized monitoring across network segments of a discrete control system, which accurately obtains the operating status data of the condensate pump station through the acquisition module, providing basic information for subsequent control and display. The introduction of the protocol conversion module effectively solves the problem of inconsistent communication protocols between different devices, enhances the compatibility and scalability of the system, and enables the system to flexibly integrate multiple devices without the need for complex modifications due to protocol differences. The water tank liquid level control module achieves precise start and stop control of the water pump based on the converted data to ensure liquid level stability, while improving energy utilization efficiency and reducing operating costs through intelligent control. The real-time display function of the display module provides operators with intuitive operating status and control execution process, facilitating timely discovery and handling of problems, and improving the reliability and maintainability of the system.

[0034] (2) The present invention realizes centralized monitoring and management throughout the entire plant through a virtual local area network and a protocol gateway, effectively solving the management problems brought about by a decentralized control system. The monitoring system can monitor the status of equipment in real time, detect and handle faults in a timely manner, and avoid environmental problems such as overflow of gas condensate pools. At the same time, with the help of the Intouch screen system, operators can remotely control the equipment, reducing the workload of manual inspections and improving management efficiency. In addition, the system reduces production costs and labor costs by reducing equipment failure rates and downtime. It also effectively avoids safety hazards such as steam leakage and condensate leakage through fault warning and alarm systems, thereby improving the safety management level of the equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to make the content of the present invention more clearly understood, the present invention is further described in detail below based on specific embodiments of the present invention in conjunction with the accompanying drawings, wherein:

[0036] Figure 1 This is a structural diagram of a cross-segment centralized monitoring system for a discrete control system in a preferred embodiment of the present invention;

[0037] Figure 2 This is a network topology diagram of a decentralized control system cross-segment centralized monitoring system in a preferred embodiment of the present invention;

[0038] Figure 3 This is a schematic diagram of a display module in a preferred embodiment of the present invention;

[0039] Figure 4 The distribution diagram of the gas condensate electric control cabinets in the whole plant in the display module in the preferred embodiment of the present invention;

[0040] Figure 5 A distribution diagram of the first partition in the discrete control system in the display module in the preferred embodiment of the present invention;

[0041] Figure 6This is a flow chart of a method for centralized monitoring across network segments of a discrete control system in a preferred embodiment of the present invention.

[0042] Explanation of the accompanying drawings in the specification: 1. Acquisition module; 2. Protocol conversion module; 3. Water tank level control module; 4. Display module; 5. PLC controller; 6. Switch; 7. Protocol gateway; 8. PLC cabinet; 9. Operation room. DETAILED DESCRIPTION

[0043] The present invention will be further described below with reference to the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it. However, the embodiments are not intended to limit the present invention.

[0044] Example 1

[0045] Reference Figure 1 As shown, the present invention provides a decentralized control system cross-segment centralized monitoring system, comprising:

[0046] Acquisition module 1, used to obtain the operating status data packet of the condensate pump station;

[0047] The protocol conversion module 2 is used to determine whether the first communication protocol of the acquisition module 1 is consistent with the preset second communication protocol; if not, the operation status data packet is converted into the second communication protocol format data and the intermediate data is obtained; otherwise, the intermediate data is obtained directly without conversion;

[0048] The water tank liquid level control module 3 is used to output a control signal for starting and stopping the water pump according to the intermediate data;

[0049] The display module 4 is used to display the parameters in the running status data packet and the execution process of the control signal in real time.

[0050] The present invention provides a centralized monitoring system across network segments of a discrete control system, which solves the problems of difficulty in centralized monitoring caused by the dispersion of existing condensate pump station systems, environmental and safety risks caused by delayed fault responses, inability to meet modern automation control requirements, low management efficiency, and delayed control responses. The acquisition module 1 can accurately obtain the operating status data of the condensate pump station, providing basic information for subsequent control and display. The introduction of the protocol conversion module 2 effectively solves the problem of inconsistent communication protocols between different devices, enhances the compatibility and scalability of the system, and enables the system to flexibly integrate multiple devices without the need for complex modifications due to protocol differences. The water tank liquid level control module 3 realizes precise start and stop control of the water pump based on the converted data, which not only ensures the stability of the liquid level, but also improves energy utilization efficiency and reduces operating costs through intelligent control. The real-time display function of the display module 4 provides operators with an intuitive operating status and control execution process, which facilitates timely discovery and handling of problems, and improves the reliability and maintainability of the system.

[0051] Specifically, refer to Figure 2 As shown, in this embodiment, virtual local area network (VLAN) technology is used to build a unified network architecture, incorporating condensate pump station equipment distributed across different sites, thereby achieving efficient and stable communication between devices across network segments. Simultaneously, with the help of protocol gateway 7, different communication protocols (such as Modbus-TCP and Profinet) are accurately converted, ensuring that various types of equipment can smoothly interact and share data, laying the communication foundation for centralized monitoring and coordinated operation of the entire system. Through the virtual local area network and protocol gateway 7, centralized monitoring and management across the entire plant are achieved, resolving the management challenges posed by decentralized control systems.

[0052] For example, the on-site discrete systems are relatively dispersed and widely exist in various areas of the factory. Based on the geographical location of the system distribution, these systems are roughly divided into 5 areas. Figure 2 , the condensate pumping station is divided into five zones, each containing different equipment and network segments, specifically:

[0053] The first zone: includes the coking control center (network segment 10.1), coking 5# gas tank (network segment 10.1), converter electrical room (network segment 10.20), and thermal power plant main control room (network segment 10.10).

[0054] The second zone: includes special steel water treatment (network segment 10.39) and 5800 blast furnace (network segment 10.45).

[0055] The third zone: includes the main power room of the fine wire workshop 3 (network segment 10.72), the secondary machine room of the converter workshop 4 (network segment 10.73), the main control room of the 135 power plant (network segment 10.73), and the blast furnace in the east area (network segment 10.22).

[0056] The fourth zone: includes the office of the second electric furnace workshop (network segment 10.29) and the air compressor station (network segment 10.10).

[0057] The fifth zone: includes the Shuixin Comprehensive Office (network segment 10.36) and the Rod and Wire Workshop 4 (network segment 10.36).

[0058] A total of 11 TCP-to-PN protocol gateways are required for 11 network segments, with 4 reserved for backup, for a total of 15. Distributed systems utilize various protocols (such as Profinet, OPC, and Modbus TCP) to enable communication and data exchange between devices and partitions, forming a complete industrial control system network.

[0059] Furthermore, in each area, the IO submodule in acquisition module 1 first collects the on-site control system's switching and analog signals. Subsequently, gateway couplers at each site are configured to efficiently communicate with the IO submodule to achieve automated control. Regarding communication protocol conversion, protocol gateway 7 is used to convert Modbus-TCP and Profinet communications. Modbus-TCP is a derivative of the Modbus family of communication protocols. It is a simple, vendor-neutral protocol specifically designed for managing and controlling automated equipment. This protocol covers the use of Modbus messages in "Intranet" and "Internet" environments using the TCP / IP protocol. Its most common use is to provide services for PLCs, I / O modules, and gateways connecting other simple domain buses or I / O modules. Based on these characteristics of the Modbus-TCP protocol, protocol gateway 7 collects signals from the underlying automated control system via the intranet Modbus-TCP protocol and then converts them to the Profinet protocol, thereby achieving physical interconnection between the various sites.

[0060] Specifically, the protocol conversion module 2 includes a data receiving submodule, a judgment submodule, and a protocol gateway 7. The protocol gateway 7 includes a data parsing submodule, a data conversion submodule, and a data sending submodule. The working principle of the various submodules included in the protocol conversion module 2 is as follows: the data receiving submodule is used to receive the operation status data packet; the judgment submodule is used to obtain the first identifier of the first communication protocol and confirm the second identifier of the preset second communication protocol; using a matching function to determine whether the first identifier and the second identifier are the same; if they are the same, the operation status data packet is directly sent to the switch 6; otherwise, a conversion signal is issued; the data parsing submodule is used to receive the conversion signal and extract valid information from the operation status data packet; the data conversion submodule converts the valid information into the second communication protocol format data according to the preset second communication protocol specifications, thereby obtaining intermediate data; and the data sending module is used to send the intermediate data to the switch 6.

[0061] Furthermore, the principle of calling a matching function includes function definition, parameter passing, conditional judgment, and return value processing. These steps together ensure that the matching function can correctly compare the two protocol identifiers and return the corresponding result.

[0062] Specifically, the core function of the protocol gateway 7 is to implement data conversion and communication between two different protocols. This is accomplished by the data parsing submodule, the data conversion submodule, and the data transmission submodule. Taking Modbus-TCP and Profinet as examples, the functions that the protocol gateway 7 can implement include: protocol conversion, data mapping, real-time communication, and device integration. The specific contents are as follows:

[0063] In terms of protocol conversion, the data conversion submodule can convert the data format of Modbus-TCP (the first communication protocol) to the data format of Profinet (the preset second communication protocol), and vice versa. This process ensures the integrity and consistency of data in different protocol environments, providing a foundation for communication between devices.

[0064] In terms of data mapping, the data conversion submodule accurately maps Modbus registers (such as holding registers and input registers) to Profinet I / O data areas. This mapping relationship enables Modbus-TCP devices and Profinet devices to find each other's data storage locations through the gateway, thereby enabling data read and write operations.

[0065] Regarding real-time communication, the data transmission submodule supports Profinet's real-time communication requirements, ensuring timely and reliable data transmission. By optimizing data transmission paths and employing efficient communication algorithms, the Protocol Gateway 7 ensures real-time and accurate data transmission, meeting the stringent real-time requirements of industrial automation production.

[0066] In terms of device integration, Protocol Gateway 7 allows Modbus-TCP devices (such as sensors and PLCs) to communicate with devices on Profinet networks (such as Siemens S7-1500 PLCs). Through the integration of Protocol Gateway 7, devices with different protocols can seamlessly connect to the same industrial automation network, enabling resource sharing and collaborative operation.

[0067] Control systems operating over vast, discrete areas often face challenges with diverse network segments and devices from different brands (using different communication protocols). This embodiment of the present invention utilizes conversion technology between two mainstream communication protocols, Modbus-TCP and Profinet, to efficiently collect signals from discrete control systems operating in different network segments and from different brands. This technology creates an architecture where multiple communication protocols coexist, enabling stable data upload and centralized remote monitoring.

[0068] Furthermore, this embodiment of the present invention uses protocol conversion to exchange data between systems. This allows data to be collected between different network segments of a discrete control system using the Modbus-TCP protocol. This data is then transmitted to the control center's main control room using virtual local area network (VLAN) technology. The data is then uploaded to the programmable controller for centralized data collection via a protocol gateway 7 using the Profinet protocol. This solution, based on the current system status and the company's actual network architecture, enables data interoperability, saves equipment investment and construction labor costs, and reduces the labor intensity of equipment maintenance.

[0069] Specifically, the relay control system in the prior art is fully upgraded to an intelligent control system based on PLC (Programmable Logic Controller). PLC is used in the water pool level control module 3 to control the operation of key equipment such as the lifting pump and the liquid level detection device. PLC is installed in each condensate pump station as the core control unit. With the help of a stable and reliable network connection, the real-time data of each pump station is efficiently transmitted to the central monitoring system. Through the PLC program, the automatic, manual, and semi-automatic control mode switching of the pump station equipment can be flexibly realized, and the seamless switching between local operation and remote control is supported, thereby improving the convenience and flexibility of operation. In addition, real-time and accurate data collection will be performed on key parameters such as water supply pressure and water pool liquid level, and these data will be intuitively presented in the display module 4. In an embodiment of the present invention, the display module 4 includes a host computer Intouch. The operator performs real-time monitoring and analysis through the visualization screen of the host computer Intouch.

[0070] Further, refer to Figure 2As shown, the PLC controller 5 and Intouch communicate via OPC (OLE for Process Control, referred to as OPC). Common OPC servers include SIMATIC NET and the OPC server officially provided by Siemens. The specific configuration steps of the OPC server are as follows:

[0071] S110: Check whether the hardware modules of the PLC controller 5 are correctly installed, including the CPU submodule and IO submodule, and ensure that all submodules are securely connected and undamaged. Complete the programming and debugging of the PLC controller 5 to ensure that the program accurately implements the intended control logic. Also, check whether the network configuration of the PLC controller 5, including the IP address and subnet mask, is correct to ensure that the PLC controller 5 can stably access the network and communicate with other devices.

[0072] S120, installing and configuring the OPC server. The OPC server serves as a key bridge connecting Intouch and the PLC controller 5, enabling efficient data interaction between the two.

[0073] Further, in Figure 2 In the example, switch 6 establishes communication between the PLC controller 5 in the pool level control module 3, the Intouch display module 4, and the protocol conversion module 2. Intouch communicates with the PLC controller 5 via the OPC protocol to read and write data. The company network connects multiple network segments via switch 6, each containing a different device system. PLC cabinet 8 houses the PLC controllers 5 for each regional site. The operating room 9 houses multiple display modules 4, which use SMC to configure data acquisition and display data on Intouch.

[0074] Further, refer to Figure 3 As shown, a display module 4 is deployed in the operating room 9. This display module 4 also includes a monitoring and data acquisition submodule, which displays the operating status of each condensate pump station in real time with high precision and frequency, covering comprehensive information such as equipment operating parameters and operating modes. Using the Intouch screen, operators can easily remotely control the start and stop of the boost pumps while simultaneously providing real-time, accurate monitoring of key parameters such as liquid level and flow rate. If the monitoring system detects abnormal data or signs of a fault, operators can quickly obtain an alert and, based on the detailed fault diagnostic data provided by the system, promptly implement targeted action, effectively shortening fault response time and ensuring the stable operation of the entire discrete control system.

[0075] Specifically, before the acquisition module 1 is run, the plant is divided into multiple areas (for example, 5 areas) based on the on-site survey, combined with the on-site equipment distribution and system equipment configuration. These areas are distributed throughout the plant and can be referred to Figures 4 and 5 As shown in the figure, each area requires independent signal acquisition and processing. Field acquisition module 1 includes an I / O submodule and a gateway coupler, which perform signal acquisition and processing. Furthermore, the selection and configuration of acquisition module 1 fully considers the brand and configuration characteristics of the field equipment.

[0076] Furthermore, the I / O submodules need to process both digital and analog signals. Different types of modules are deployed on-site, such as digital I / O modules and analog I / O modules. Given the high demands placed on accurate and real-time signal acquisition, especially in industrial environments where electromagnetic interference may be a concern, module selection prioritizes the module's protection level and anti-interference capabilities. For example, modules with an IP67 protection rating are preferred to ensure stable operation in harsh industrial environments.

[0077] Furthermore, the gateway coupler is responsible for aggregating and transmitting data from each I / O submodule to the tank level control module 3. The choice of communication protocol is particularly critical. Common protocols include Profinet, Modbus TCP, and EtherNet / IP. The specific choice will be determined by the existing decentralized control system architecture. Furthermore, the processing power and stability of the gateway are also crucial, especially in a distributed system with multiple zones (such as five). The network topology may require redundant designs, such as ring or star topologies, to ensure that a node failure does not affect overall communication.

[0078] Furthermore, regarding automated control, the PLC controller 5 performs data processing and logic control for device signal acquisition. Because equipment locations are relatively dispersed, each area requires local signal acquisition and logic control to reduce the burden on the PLC controller 5. Consequently, the PLC controller 5 at each site is able to perform real-time data processing and feedback, while also transmitting important data via a gateway coupler to the tank level control module 3 and display module 4 for monitoring and analysis.

[0079] Furthermore, regarding network architecture, assuming the site is divided into five equipment distribution areas, existing industrial Ethernet and fiber optic communication networks, or a combination of both, can be utilized locally. Fiber optic communication can effectively ensure the stability of long-distance transmission, especially in large factories with long distances.

[0080] Furthermore, in terms of integration of the monitoring system described in the embodiment of the present invention, the OPC UA protocol can effectively solve compatibility issues between different devices, facilitate data uploading to the InTouch monitoring platform, and achieve efficient system integration.

[0081] Furthermore, reliability design incorporates redundant configurations, such as dual power supplies and dual network links, to prevent single points of failure. Furthermore, lightning protection and grounding are crucial in a factory environment, especially for distributed systems, which may be exposed to greater environmental risks. Furthermore, the on-site gas condensate equipment is located in the gas area, so an explosion-proof control cabinet is installed to complete the system and ensure safety.

[0082] The embodiments of the present invention advance project implementation in phases, starting with a pilot program and then gradually expanding, based on the usage of on-site equipment. This step-by-step implementation approach allows for preliminary testing of the solution's effectiveness on a small scale, enabling timely identification and resolution of issues, laying a solid foundation for subsequent full-scale rollout. During the debugging and optimization phase, issues such as signal delays and packet loss were encountered. To address these issues, a series of measures were implemented, including network stress testing and signal transmission optimization, to ensure stable system operation.

[0083] Specifically, the monitoring system described in the embodiment of the present invention also integrates an early warning module for real-time monitoring of the parameters of the operating status data packet, comparing the parameters with the alarm threshold, determining whether there is an abnormality, and promptly issuing an alarm signal when an abnormality is detected. Specifically, when an abnormal fluctuation in the liquid level or a failure of the equipment is detected, the module will quickly and automatically trigger the alarm mechanism, immediately reminding the operator to intervene and handle the problem in a timely manner. The alarm method is flexible and diverse, covering various forms such as text messages, emails, and sound and light alarms, ensuring that fault information can be quickly and accurately conveyed to the relevant responsible persons, thereby ensuring that the system can respond in a timely manner and effectively handle emergencies.

[0084] Specifically, the water tank level control module 3 also has an automatic recording function, which is mainly implemented by the recording submodule. In addition, the water tank level control module 3 also includes a data preprocessing submodule, a feature extraction submodule, a prediction submodule and an output submodule. The working principles of each submodule in the water tank level control module 3 are as follows:

[0085] The recording submodule is used to collect and store intermediate data and generate historical data according to preset rules; the data preprocessing submodule is used to clean, normalize and time series align the historical data to obtain the first data; the feature extraction submodule is used to extract features from the first data to obtain the liquid level change law characteristics and the water pump start-stop law characteristics; based on the liquid level change law characteristics and the water pump start-stop law characteristics, deep feature mining and data correlation analysis are fused to obtain a feature set; the prediction submodule is used to construct a liquid level prediction model and train the liquid level prediction model based on the feature set to obtain the liquid level prediction result; the output submodule is used to output the control signal for controlling the start and stop of the water pump based on the liquid level prediction result.

[0086] Furthermore, the preset rules include time dimension rules, data storage capacity rules, and data validity rules. Time dimension rules include sampling at fixed time intervals and concentrated sampling at specific time periods. Data storage capacity rules include limiting the amount of stored data and dynamically adjusting the sampling frequency based on storage space. Data validity rules include data range limitations and data continuity checks. By properly setting these preset rules, the recording submodule can efficiently and accurately collect and store historical data that meets the requirements, providing a solid data foundation for the subsequent operations of the various submodules of the pool liquid level control module 3.

[0087] For example, for fixed-interval sampling, set the tank level data to be collected every 5 minutes. This ensures data continuity and a certain density, facilitating subsequent analysis of short-term level fluctuations. For example, starting at 8:00 AM, record the level value every 5 minutes, such as once at 8:00 AM, once at 8:05 AM, and so on.

[0088] For example, for centralized sampling based on specific time periods, liquid level data is collected every 2 minutes during peak water usage hours (e.g., 6:00 AM to 9:00 AM and 6:00 PM to 9:00 PM) and every 10 minutes during other hours. Because liquid levels can change rapidly during peak hours, more intensive data is needed to capture the details of these changes. During off-peak hours, liquid levels change relatively slowly, so appropriately reducing the sampling frequency can reduce the amount of data stored while still meeting basic monitoring needs.

[0089] For example, to limit the amount of stored data, the recording submodule is set to store a maximum of 1,000 historical data items. When new data is collected and the storage capacity reaches 1,000 items, the oldest data item is deleted according to the first-in, first-out (FIFO) principle to make room for the new data. For example, if the earliest data item collected is item 1, when item 1001 is collected, the first item is automatically deleted and the 1001st item is stored.

[0090] For example, when dynamically adjusting the sampling frequency based on storage space, if the remaining storage capacity is detected to be less than 20%, the sampling frequency will be automatically reduced from every 5 minutes to every 15 minutes until storage space is increased or data is cleared. For example, if data is originally collected every 5 minutes, when storage space is tight, it will be reduced to every 15 minutes, reducing the amount of data stored and alleviating storage pressure.

[0091] For example, for data range limitation, the valid range of the pool level data is set to 0-5 meters (assuming the maximum depth of the pool is 5 meters). If the collected level data exceeds this range (such as a negative value or a value greater than 5 meters), the data is marked as invalid and the abnormality is recorded. For example, if the collected level value is -0.2 meters, the data is recorded as invalid and an abnormality alarm is issued.

[0092] For example, for data continuity checks, if the change in liquid level data collected for five consecutive times exceeds a preset threshold (such as 1 meter), the data is considered to be abnormal and marked as suspicious. For example, under normal circumstances, the liquid level changes are relatively stable, but suddenly the liquid level jumps from 3 meters to 4 meters, 5 meters, 6 meters, 7 meters, and 8 meters for five consecutive times. This situation does not conform to the normal liquid level change pattern and the data is marked as suspicious.

[0093] The recording submodule can accurately collect and store various types of data during the operation of the equipment, and then generate detailed historical data reports based on preset rules. These reports not only provide a data basis for subsequent in-depth analysis, but also help optimize production processes and equipment management strategies. Based on in-depth analysis of massive data, the system builds a set of intelligent management models. This model can automatically and accurately control the liquid level based on real-time data, predict the optimal start and stop time of the pump group in advance, and effectively prevent the occurrence of liquid level overflow accidents; at the same time, through continuous monitoring and analysis of equipment operation data, the model can accurately predict possible equipment failures, thereby arranging maintenance work in advance, reducing equipment downtime, reducing maintenance costs, and improving equipment reliability and operating efficiency.

[0094] Specifically, the specific steps for the water tank liquid level control module 3 to predict the optimal start and stop timing of the pump group in advance are:

[0095] S210 uses sensors such as liquid level sensors, current / voltage sensors, and temperature / vibration sensors to collect key operating data from the condensate pump station in real time and transmit this data to the PLC (Programmable Logic Controller) via a communication link. This data covers important parameters such as the water tank liquid level, pump operating status, current and voltage levels, and equipment temperature and vibration, providing a data foundation for subsequent intelligent management.

[0096] S220, the PLC, serves as the core control unit and processes the operational data transmitted by acquisition module 1. Based on pre-set control logic, the PLC can control the start, stop, and switching of the water pump. For example, when the liquid level sensor detects that the pool level has reached the preset high threshold, the PLC automatically starts the pump to drain the water. When the level drops to the low threshold, the PLC stops the pump, ensuring that the pool level remains within a safe range.

[0097] Furthermore, in the pool level control module 3, an intelligent management model (i.e., a prediction model) based on data analysis was constructed. This model can automatically control the liquid level and accurately predict the start and stop timing of the pump group, effectively preventing liquid overflow accidents. The overall framework is based on data analysis and machine learning models, and uses intelligent means to achieve accurate prediction and control of liquid levels. The specific steps are as follows:

[0098] S221. Collect key data such as pool level, flow, and pump operating status in real time to ensure the timeliness and accuracy of the data.

[0099] S222, the data preprocessing submodule cleans, normalizes, and time-series aligns the collected raw data. The cleaning process includes removing noise data and outliers, and normalization converts data of different dimensions into a unified format for subsequent analysis and processing. Time series data alignment ensures the continuity and consistency of the data in the time dimension, providing high-quality data support for subsequent feature extraction and model training. The specific steps of data preprocessing are:

[0100] Step 1: Data cleaning. The collected raw data is cleaned to remove outliers and missing values. Outliers can be identified by setting a reasonable threshold range; data points outside this range are considered outliers. Missing values ​​can be filled using methods such as interpolation to ensure data integrity.

[0101] Step 2: Normalization. Map the liquid level, flow rate and other data to the range of [0,1] through normalization. The normalization formula is:

[0102]

[0103] Among them, x is the original data, x max and x min are the minimum and maximum values ​​in the data sequence, respectively, norm is the normalized data value.

[0104] Step 3: Time series alignment. Ensure that data such as liquid level, flow rate, and pump status are strictly aligned in the time dimension. Specifically, adjust each data series according to a unified timestamp so that they have corresponding data values ​​at the same time point, providing accurate time series data for subsequent feature extraction and model training.

[0105] S223: The feature extraction submodule extracts key features from the first data, such as the pattern of liquid level changes and the water pump start-stop pattern. By analyzing the first data, this submodule can identify the periodicity and trend of liquid level changes, as well as the correlation between liquid level changes and water pump start-stop operations. These features are the foundation for the machine learning model to make accurate predictions and help the model better understand the inherent patterns of liquid level changes.

[0106] Exemplarily, the following features are extracted from the first data:

[0107] Liquid level change rate: The rate of change of liquid level over time. The calculation formula is:

[0108]

[0109] Among them, h t is the current liquid level, h t-1 is the liquid level at the previous moment, and Δt is the time interval.

[0110] Relationship between flow and liquid level: The influence of flow on liquid level is calculated as follows:

[0111] Q net =Q in -Q out ;

[0112] Among them, Q in is the water inlet flow rate, Q out is the water outflow.

[0113] Water pump running time: the relationship between the water pump running time and the change of liquid level.

[0114] Furthermore, the process of integrating deep feature mining and data association analysis includes establishing association rules between liquid level data and related data, and calculating the support and confidence of the association rules. The related data includes equipment failure data and environmental factor data.

[0115] For example, the regular characteristics of liquid level changes include the liquid level change rate, the liquid level sliding average, and the liquid level standard deviation; the regular characteristics of water pump start and stop include the number of water pump starts and stops, the continuous operation time, and the start and stop interval time. The regular characteristics of liquid level changes, the regular characteristics of water pump start and stop, and the depth characteristics are analyzed for correlation with relevant data (such as equipment failure data and environmental factor data). An association rule mining algorithm (such as the Apriori algorithm) is used to discover potential association relationships. The specific association rules are:

[0116] Rule 1: If the liquid level change rate exceeds threshold T1, and the number of pump starts and stops in the last hour exceeds threshold T2, then fault event F1 exists in the equipment fault data.

[0117] Rule 2: If the liquid level sliding average is lower than threshold T3 and the temperature in the environmental factor data exceeds threshold T4, the pump start-stop interval will increase significantly.

[0118] Rule 3: If the depth feature di exceeds the threshold T5 and the liquid level standard deviation is lower than the threshold T6, the liquid level changes smoothly and the pump is in good operating condition.

[0119] For Rule 1, its support and confidence are calculated as follows: For support, the frequency of the conditions that the liquid level change rate exceeds T1, the number of pump starts and stops exceeds T2, and the fault event F1 exists simultaneously in all data. For confidence, the probability of the fault event F1 existing when the liquid level change rate exceeds T1 and the number of pump starts and stops exceeds T2 is calculated.

[0120] For Rule 2, its support and confidence are calculated as follows: In terms of support, the frequency of the simultaneous occurrence of the liquid level sliding average below T3, the temperature exceeding T4, and a significant increase in the pump start-stop interval across all data sets. In terms of confidence, the probability of a significant increase in the pump start-stop interval when the liquid level sliding average below T3 and the temperature exceeding T4 are both met.

[0121] For Rule 3, the support and confidence are calculated as follows: For support, the frequency of all data simultaneously meeting the conditions that the depth feature di exceeds T5, the liquid level standard deviation is lower than T6, and the liquid level changes smoothly and the pump is operating well. For confidence, the probability that the liquid level changes smoothly and the pump is operating well is determined when the depth feature di exceeds T5 and the liquid level standard deviation is lower than T6.

[0122] By combining the characteristics of liquid level variation patterns, pump start-stop patterns, and deep features, and performing data association analysis, a rich feature set was generated. This feature set not only includes traditional statistical features but also incorporates complex patterns extracted through deep learning and the results of association rule analysis. This provides more comprehensive and accurate input data for subsequent liquid level prediction and pump control model training, thereby improving the model's predictive performance and control accuracy.

[0123] In the prediction submodule, S224, a machine learning model is used to train the extracted features to predict future liquid level trends. Through extensive training with historical data, the model learns the complex patterns of liquid level changes and predicts future level changes based on the current operating status, providing a basis for intelligent control decisions.

[0124] Specifically, in the embodiment of the present invention, the machine learning models that can be used include linear regression model, LSTM (Long Short-Term Memory Network), random forest regression, etc. The machine model can be selected according to actual needs.

[0125] For example, for the linear regression model, it is assumed that there is a linear relationship between the liquid level change and the flow rate and the pump status, and a linear regression model is constructed. The prediction formula of the model is:

[0126] h t+1 =w1·h t +w2·Q net +w3·S;

[0127] Where: h t+1 is the predicted value of the liquid level at the next moment, h t is the current liquid level; Q net is the net flow; S pump is the pump status (0 means stopped, 1 means running); w1, w2, and w3 are model weights, which are used to adjust the influence of each feature on the liquid level change; b is the bias term, which is used to adjust the overall offset of the model.

[0128] For example, the LSTM (Long Short-Term Memory) deep learning model used to process time series data is suitable for capturing long-term dependencies in liquid level changes. Its core structure includes a forget gate, input gate, candidate memory, memory unit, output gate, etc., which can effectively solve the gradient vanishing problem in traditional recurrent neural networks (RNNs). The update formula of LSTM is:

[0129] f t =σ(W f ·[h t-1 ,x t ]+b f );

[0130] i t =σ(W i ·[h t-1 ,x t ]+b i );

[0131]

[0132] O t =σ(W o ·[h t-1 ,x t ]+b o );

[0133] h t =O t tanh(C t );

[0134] Among them, f t is the forget gate, which controls the retention degree of the memory at the previous moment; i t It is the input gate, which controls the acceptance level of the current input; is the candidate memory; C t is the current memory; O t is the output gate, which controls the output memory; h t is the current output; σ is the Sigmoid activation function; W f 、W i 、W C 、b f、b i 、b C are different parameters of the model.

[0135] For example, for random forest regression, random forest is an integrated learning method suitable for nonlinear relationship modeling. Its prediction formula is:

[0136]

[0137] Where: f i (x) is the prediction result of the i-th decision tree; N is the number of decision trees.

[0138] S225, the output submodule, automatically controls the start and stop of the water pump based on the model's predictions. If the liquid level is predicted to reach the high threshold, the pump is started to drain the water; if the liquid level is predicted to fall to the low threshold, the pump is stopped immediately. This predictive intelligent control approach effectively prevents overflow accidents while improving system efficiency and reliability.

[0139] Specifically, based on the prediction results, the control strategy is:

[0140] Liquid level threshold control satisfies: when h t+1 >h max Start the water pump; when h t+1 <h max Stop the water pump; where h max and h min The upper and lower limit thresholds of the liquid level.

[0141] Predictive control: Based on the liquid level prediction results, the water pump is started and stopped in advance to prevent the liquid level from exceeding the threshold.

[0142] S226, the water tank liquid level control module 3 also includes a feedback submodule, which is used to monitor the actual changes in the water tank liquid level in real time, evaluate the degree of deviation in the current liquid level control, and dynamically adjust the prediction model. Specifically, through real-time data feedback, the machine learning model is continuously optimized. During actual operation, new operating data is continuously collected and fed back into the model for retraining and optimization. This dynamic feedback mechanism ensures that the model is always in optimal condition and adapts to changes in the system operating environment, thereby continuously improving the system's intelligence level and control accuracy.

[0143] Specifically, in the process of machine learning model training and optimization, the loss function is preferably the mean square error (MSE), whose mathematical expression is:

[0144]

[0145] Among them, N1 is the number of samples, hpred is the actual value of the i-th sample, h true The predicted value of the i-th sample.

[0146] Furthermore, during the model training process, optimization algorithms such as gradient descent and Adam optimizer can be used to update model parameters to ensure that the model can efficiently converge to the optimal solution.

[0147] S230. Integrate the trained machine learning model into the control system of the condensate pump station to implement the following functions: real-time monitoring, real-time collection of data such as liquid level, flow rate, and pump status; automatic control, automatically starting and stopping the pump based on the prediction results; an alarm mechanism, triggering an alarm when the liquid level exceeds the safe range; model update, regularly retraining the model with new data to improve prediction accuracy.

[0148] S240: Based on the above steps, the host computer monitoring screen in the display module 4 is used to display important information such as the water tank liquid level, water pump operating status, and equipment parameters in real time. Through the visual interface, the operator can clearly understand the operating status of the entire system, making it easier to identify and address problems in a timely manner.

[0149] Steps S210 to S240 above construct an intelligent liquid level control system through the deep integration of data analysis and machine learning models. This system not only enables automatic liquid level control and precise pump startup and shutdown, but also effectively prevents liquid overflow accidents, improving control accuracy and operational efficiency. Furthermore, the system's intelligent pre-judgment capabilities and comprehensive alarm mechanism further enhance its reliability and safety, providing a strong guarantee for the safe and efficient operation of the condensate pumping station.

[0150] Furthermore, the tank level control module 3 also includes a data analysis submodule that conducts in-depth analysis and prediction of the equipment's operating status. This not only provides equipment lifespan predictions but also provides early warnings of faults. This helps plan equipment maintenance in advance, reduces the impact of sudden failures on production, reduces maintenance costs, and improves equipment reliability and operational efficiency. When the system detects an abnormality, in addition to triggering an alarm to alert the operator, it also automatically records detailed log information. These logs record the time, cause, and measures taken for the abnormality, providing a powerful basis for equipment management and event tracing, facilitating subsequent analysis and improvement.

[0151] Furthermore, the control system configuration of this embodiment of the present invention utilizes a regional data collection and centralized control model. This model supports distributed data collection of over 5,000 data points, ensuring that the system control cycle does not exceed 10 milliseconds and meets real-time requirements. Leveraging advanced multi-protocol fusion technology, regional data synchronization accuracy can be controlled to less than 1 microsecond. This method is expected to reduce wiring costs by 40% while significantly improving system reliability to 99.999%.

[0152] The embodiment of the present invention achieves centralized monitoring and management throughout the entire plant through a virtual local area network and protocol gateway 7, effectively resolving the management challenges posed by distributed control systems. The automated monitoring system can monitor equipment status in real time, promptly detect and address faults, and avoid environmental issues such as overflow of gas condensate pools. At the same time, with the help of the Intouch screen system, operators can remotely control equipment, reducing the workload of manual inspections and improving management efficiency. Furthermore, the system reduces production and labor costs by reducing equipment failure rates and downtime. Furthermore, through the fault warning and alarm system, it effectively avoids safety hazards such as steam leaks and condensate leaks, thereby improving the safety management level of the equipment and enhancing the overall safety of the system.

[0153] Example 2

[0154] Based on the same inventive concept, this embodiment provides a method for centralized monitoring of a discrete control system across network segments. The principle of solving the problem is similar to that of a centralized monitoring system for a discrete control system across network segments provided in Example 1, and the repeated parts will not be repeated.

[0155] Reference Figure 6 As shown, the present invention provides a method for centralized monitoring of a discrete control system across network segments, including but not limited to the following steps:

[0156] Obtain the operating status data packet of the condensate pump station;

[0157] Determine whether the first communication protocol for transmitting the running status data packet is consistent with the preset second communication protocol; if not, convert the running status data packet into data in the second communication protocol format and obtain intermediate data; otherwise, directly obtain the intermediate data without conversion;

[0158] According to the intermediate data, the control signal for starting and stopping the water pump is output;

[0159] Monitor parameter changes in the operating status data packet and the execution process of the control signal in real time.

[0160] Example 3

[0161] This embodiment provides a cross-segment centralized monitoring device for a discrete control system, including a cross-segment centralized monitoring system for a discrete control system provided in the first embodiment.

[0162] The monitoring device described in the present invention ensures high reliability and stable operation in harsh environments by adopting industrial-grade hardware and software; by optimizing the control logic, it achieves energy saving and high efficiency, reduces energy consumption and extends the service life of the equipment; the friendly monitoring interface enhances operability, allowing operators to conveniently monitor and manage the system; at the same time, the monitoring device has good scalability and supports functional expansion, which is convenient for future upgrades and modifications; in addition, through data analysis and prediction functions, the monitoring device realizes intelligent control and management of equipment status, thereby improving the overall intelligence level.

[0163] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0164] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0165] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0166] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0167] Obviously, the above embodiments are merely examples for clarity of explanation and are not intended to limit the implementation methods. Those skilled in the art will appreciate that other variations or modifications can be made based on the above description. It is not necessary and impossible to enumerate all implementation methods here. Obvious variations or modifications arising therefrom remain within the scope of protection of the present invention.

Claims

1. A decentralized control system cross-segment centralized monitoring system, characterized by: include: The acquisition module is used to obtain the operating status data packet of the condensate pump station; a protocol conversion module, configured to determine whether the first communication protocol of the acquisition module is consistent with a preset second communication protocol; if not, converting the running status data packet into data in the second communication protocol format and obtaining intermediate data; otherwise, obtaining the intermediate data directly without performing the conversion; A water tank liquid level control module, configured to output a control signal for starting and stopping the water pump according to the intermediate data; The display module is used to display the parameters in the operation status data packet and the execution process of the control signal in real time.

2. A decentralized control system cross-segment centralized monitoring system according to claim 1, characterized in that , the water tank liquid level control module includes: A recording submodule, for collecting and storing the intermediate data and generating historical data according to preset rules; A data preprocessing submodule, configured to clean, normalize, and time-series-align the historical data to obtain first data; a feature extraction submodule for extracting features from the first data to obtain liquid level variation regularity features and water pump start-stop regularity features; and performing fusion deep feature mining and data correlation analysis based on the liquid level variation regularity features and water pump start-stop regularity features to obtain a feature set; A prediction submodule is used to construct a liquid level prediction model and train the liquid level prediction model according to the feature set to obtain a liquid level prediction result; The output submodule is used to output a control signal for controlling the start and stop of the water pump according to the liquid level prediction result.

3. A decentralized control system cross-segment centralized monitoring system according to claim 2, characterized in that ,The process of integrating deep feature mining and data association analysis in the ,feature extraction submodule includes establishing association rules between liquid level data and ,related data; and calculating the support and confidence of the ,association rules.

4. A decentralized control system cross-segment centralized monitoring system according to claim 2, characterized in that ,The water pool level control module also includes a feedback submodule for real-time monitoring of the actual changes of the water pool level, evaluating the degree of deviation of the current level control, and then dynamically adjusting the prediction model.

5. The cross-segment centralized monitoring system for a discrete control system according to claim 1, characterized in that: The protocol conversion module includes: A data receiving submodule, configured to receive the running status data packet; The judgment submodule is used to obtain the first identifier of the first communication protocol and confirm the second identifier of the preset second communication protocol; use a matching function to determine whether the first identifier and the second identifier are the same; if they are the same, directly send the operating status data packet to the switch; otherwise, send a conversion signal.

6. A decentralized control system cross-segment centralized monitoring system according to claim 1 or 5, characterized in that: The protocol conversion module also includes: A data parsing submodule, configured to receive the conversion signal and extract valid information from the operation status data packet; A data conversion submodule converts the valid information into data in the format of the second communication protocol according to the specification of the preset second communication protocol, and obtains intermediate data; The data sending submodule is used to send the intermediate data to the switch.

7. The cross-segment centralized monitoring system for a discrete control system according to claim 1, characterized in that: It also includes a switch for establishing communication between the water pool liquid level control module, the display module and the protocol conversion module.

8. The cross-segment centralized monitoring system for a discrete control system according to claim 1, characterized in that: It also includes an alarm module for monitoring the parameters of the running status data packet in real time, comparing the parameters with the alarm threshold, judging whether there is an abnormality, and issuing an alarm signal in time when an abnormality is detected.

9. A method for centralized monitoring of a discrete control system across network segments, characterized in that: include: Obtain the operating status data packet of the condensate pump station; Determining whether the first communication protocol for transmitting the operating status data packet is consistent with a preset second communication protocol; If they are inconsistent, converting the running status data packet into data in the second communication protocol format and obtaining intermediate data; otherwise, not converting and directly obtaining the intermediate data; Outputting a control signal for starting and stopping the water pump according to the intermediate data; The parameter changes in the operating status data packet and the execution process of the control signal are monitored in real time.

10. A centralized monitoring device for a discrete control system across network segments, characterized in that: It comprises a decentralized control system cross-segment centralized monitoring system as described in any one of claims 1 to 8.

Images