A desulfurization DCS intelligent voice alarm and custom state display system
Patent Information
- Application Number
- CN202611102586.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-23
- Publication Date
- 2026-09-25
AI Technical Summary
二期脱硫系统覆盖吸收塔系统、浆液制备系统、烟气系统、公用辅助系统等多个工艺单元,现场测控测点数量多、设备连锁逻辑复杂,在2024年9月改造后首次168小时试运期间,原有报警系统暴露出严重缺陷:单台浆液循环泵跳闸常引发20余条连锁报警同时涌入,运行人员平均需耗时约20秒才能从报警列表中人工分辨出根源事件;同时,由于报警显示样式为组态阶段固化预设,不同班次人员为适应视觉习惯,多次要求工程师站停机修改参数,严重影响监控连续性
一、本发明针对脱硫系统工艺连锁链路长、单点故障易引发批量次生报警的特点,采用带关联抑制的动态优先级播报算法,基于脱硫工艺连锁拓扑区分根源报警与次生连锁报警,结合报警等级、工艺影响程度、机组负荷、设备运行状态多维度计算播报权重,从根源上缓解报警风暴对运行人员的干扰,确保核心故障信息优先播报,有效缩短脱硫系统故障定位与处置的响应时长,提升机组环保设施运行的稳定性。
Smart Images

Figure CN122821701A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of flue gas desulfurization distributed control systems for thermal power plants, in particular to a desulfurization DCS intelligent voice alarm and custom status display system. Background Art
[0002] As a core environmental protection facility supporting the unit, the phase II flue gas desulfurization unit of Qingdao Power Plant realizes its operation control based on a distributed control system, and the system is currently undergoing localization transformation and upgrading. The phase II desulfurization system covers multiple process units including an absorber tower system, a slurry preparation system, a flue gas system, and a public auxiliary system. It has a large number of on-site measurement and control points and complex equipment interlocking logic. During the first 168-hour trial operation after the transformation in September 2024, the original alarm system exposed serious defects: the tripping of a single slurry circulating pump often triggers more than 20 interlocking alarms to flood in simultaneously, and operators need an average of about 20 seconds to manually identify the root cause from the alarm list; meanwhile, since the alarm display style is solidified and preset during the configuration stage, operators of different shifts have repeatedly requested the engineer station to shut down and modify parameters to adapt to their visual habits, which seriously affects the continuity of monitoring. In daily operation, the DCS alarm system is fully relied on to realize process abnormality early warning, equipment fault prompt and protection action feedback. With the gradual application of domestic CPUs and domestic operating systems in DCS, the original supporting alarm system can no longer meet the actual requirements of on-site refined operation management in terms of functional adaptability and operation and maintenance flexibility.
[0003] Traditional desulfurization DCS alarm systems adopt a sequential broadcasting mode with fixed levels, and do not distinguish between root faults and secondary interlocking alarms. A single core equipment fault can easily trigger dozens of related alarms to broadcast simultaneously, forming an alarm storm that interferes with operators to quickly locate the core fault point; the voice broadcasting strategy is fixed, and cannot dynamically adjust the broadcasting priority in combination with unit load and equipment operation status, so key alarm information is easily submerged by a large number of secondary alarms under high load conditions. The traditional alarm display style is a fixed mode preset in the configuration stage, and operators cannot adjust display parameters online according to their own operating habits, resulting in poor operational adaptability for operators of different shifts. Adjusting the display configuration requires switching to the engineer configuration mode, which interrupts normal operation monitoring. Meanwhile, the traditional alarm system only collects alarms from process-side measuring points, lacks alarms for the operation status of DCS's own hardware nodes and core processes, and has insufficient linkage between alarm data and monitoring screens, logic configuration, and historical trends. Fault troubleshooting requires manual searching for corresponding measuring points across modules, resulting in low operation and maintenance troubleshooting efficiency.
[0004] In summary, existing desulfurization DCS alarm systems suffer from insufficient differentiation in broadcast priority, poor flexibility in display configuration, lack of system self-monitoring capabilities, and weak cross-module data linkage. They cannot meet the safe and efficient operation requirements of desulfurization systems after domestic modification. Therefore, it is necessary to develop a desulfurization DCS intelligent voice alarm and customized status display system that is compatible with domestic software and hardware platforms to improve the operation monitoring level and fault handling capabilities of desulfurization systems. Summary of the Invention
[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a desulfurization DCS intelligent voice alarm and customized status display system. This system can synchronously collect desulfurization process monitoring point alarms, equipment displacement signals, and DCS node operating status data through an alarm data source acquisition module, and transmit them synchronously to an intelligent voice alarm processing module and an alarm status customized display module. It calculates broadcast priority using a dynamic weighted algorithm with correlation suppression, distinguishes between primary and secondary alarms, and dynamically sorts the broadcasts based on unit load. An online configuration and real-time effect of alarm display parameters are achieved through a shared memory architecture in operation mode. A matching alarm linkage positioning unit enables rapid jumps from alarm points to screens, configurations, and trends. Overall, it realizes intelligent hierarchical broadcasting and personalized display of desulfurization DCS alarms, improving the operation and maintenance efficiency and fault response speed of the desulfurization system.
[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a desulfurization DCS intelligent voice alarm and customized status display system, which includes the following components: Alarm data source acquisition module: It connects to the desulfurization DCS process control station and the field measurement and control equipment. The output end is connected to the input end of the intelligent voice alarm processing module and the alarm status custom display module respectively. The alarm data source acquisition module synchronously collects the desulfurization process measurement point over-limit signal, equipment change signal, protection action signal and the corresponding equipment operation status label data. The equipment operation status label is obtained by combining the contactor auxiliary contact collected by hard wiring and the DCS internal sequential control step sequence. Intelligent voice alarm processing module: It has a built-in priority calculation unit and an offline voice synthesis unit. It dynamically calculates the broadcast priority of each alarm based on the alarm level, process impact degree, equipment operating status, and unit load parameters. It executes the corresponding mode of voice broadcast according to the priority order. The priority calculation is based on the root cause and secondary determination of the pre-stored desulfurization process interlocking topology relationship map. This map is stored in the non-volatile memory of the process control station in the form of an adjacency list. Alarm status custom display module: It has a built-in online configuration unit and real-time rendering unit. It receives the display parameters input by the operator and performs legality verification. The parameters that pass the verification are written into the running cache and take effect directly without restarting the system or reloading the project files.
[0007] Furthermore, the offline speech synthesis unit of the intelligent voice alarm processing module has a built-in corpus of desulfurization professional terminology. The corpus covers professional terms and alarm descriptions for the entire process of desulfurization tower, slurry circulation pump, flue gas heat exchanger, absorbent preparation, and wastewater treatment. All professional terms have been calibrated for pronunciation. The module has a built-in environmental noise acquisition unit that collects the ambient noise value in dB in real time through the external pickup device of the operator station. The volume of the voice broadcast is dynamically adjusted according to the noise value. For every 10 dB increase in noise, the broadcast volume is increased by 15% relative to the system reference volume. The volume adjustment range is 30%-100%. The module supports setting a silent period. During the silent period, only the broadcast function of the highest level alarm is retained. Other levels of alarms are only triggered to display prompts. The module supports custom broadcast intervals. The interval time for the same repeated broadcast can be set in the range of 30s-300s. The interval length matches the normal change rate of desulfurization process parameters to avoid information interference caused by frequent broadcasts.
[0008] Furthermore, the priority calculation unit uses a dynamic weighted algorithm with correlation suppression to calculate the broadcast priority score of a single alarm. The formula is: ,in The alarm levels are based on a base score of 100, 70, 40, and 15 points respectively, with the four levels being Emergency, Important, General, and Warning. The level classification is determined according to the degree of impact of desulfurization system malfunctions on unit operation and environmental emissions. This is the process influence coefficient, ranging from 0.1 to 1.0. It is pre-calibrated based on the impact range of a fault in the subsystem to which the measuring point belongs. Specific calibration values are: 1.0 for the core subsystem (absorber level, flue gas outlet damper); 0.7 for important subsystems (slurry circulation pump, oxidation fan, absorbent preparation equipment); and 0.4 for auxiliary process measuring points such as wastewater treatment and sump pumps. This is the unit load correction factor, with values set according to the power plant load dispatch curve: 1.2 when the unit load is above 90% of the rated load, 1.0 when the load is between 50% and 90%, and 0.8 when the load is below 50%. This is the equipment operating status coefficient. An alarm value of 1.0 is assigned to equipment in operation, and an alarm value of 0 is assigned to equipment in shutdown. This is the alarm correlation suppression coefficient, ranging from 0.4 to 1.0. Specifically, it is set to 1.0 for root cause alarms, 0.6 for primary secondary cascading alarms, and 0.4 for secondary and higher-level secondary cascading alarms. The correlation relationship is determined based on pre-stored desulfurization process cascading topology. , , The weighting coefficients are 0.6, 0.2, and 0.2, respectively. These weighting coefficients were calibrated after consistency verification using the analytic hierarchy process (AHP) combined with statistical data on desulfurization failures from our plant over the past three years. All alarms are set according to... Alarms are broadcast in descending order of priority. Alarms of the same priority are broadcast in the order of their trigger time. During the broadcast, newly triggered high-priority alarms are automatically inserted at the beginning of the current broadcast queue, while low-priority alarms are automatically delayed. Alarms that have been broadcast in the queue are automatically removed. Unconfirmed emergency level alarms are repeatedly entered into the broadcast queue at the set broadcast interval.
[0009] Furthermore, the intelligent voice alarm processing module incorporates a condition-adaptive alarm suppression unit. This unit includes built-in equipment operation status association rules. It determines the equipment status based on the equipment hard-contact operation feedback signals collected by the process control station. When the equipment is in a shutdown, maintenance, or isolation state, it automatically blocks all alarms corresponding to that equipment. The blocking state switches in real time according to the equipment operation feedback signals, without the need for manual settings. The module supports setting alarm delay triggering functions, allowing for trigger delays of 0.5s-10s for process measuring points with different fluctuation characteristics. For measuring points with rapid fluctuations in liquid level and pressure, the delay is set to 0.5s-2s, while for measuring points with gradual changes in pH and concentration, the delay is set to 3s-10s. A formal alarm is only triggered after the measuring point exceeds the delay threshold, avoiding false alarms caused by fluctuations in the measuring point signals. For alarms caused by interlocking protection actions in the desulfurization system, the module automatically classifies them as emergency level, immediately inserts the voice broadcast at the top of the queue, and triggers a full-screen pop-up prompt. The pop-up content includes the name of the protection action, the action time, and a list of associated equipment. The pop-up requires manual confirmation from the operator before it can be closed.
[0010] Furthermore, the online configuration unit of the alarm status custom display module provides a visual interactive interface. Operators can directly adjust alarm window parameters through the operator station interface. The adjustments include the font color, background color, flashing frequency, and text size of alarm items, as well as the docking position, display size, and number of items displayed per page of the alarm window. All adjustments do not require entering the engineer configuration mode. After the parameters are input, the configuration verification unit performs a legality check, which includes the parameter value range and format compatibility. Parameters that pass the check are directly written to the running shared memory area. The real-time rendering unit reads the configuration parameters in the shared memory at a fixed period of 100ms and refreshes the alarm display interface. The parameter modification process is smooth without screen lag or data interruption. All monitoring interfaces on the same operator station share the same shared memory area, and the configuration parameters are automatically synchronized to all open alarm windows and monitoring screens, avoiding the problem of inconsistent display across multiple interfaces.
[0011] Furthermore, the alarm status custom display module has a built-in configuration scheme storage unit that supports saving up to 8 independent display schemes. The number of schemes matches the power plant's four-shift, three-operation shift configuration plus two emergency backup schemes. Each scheme corresponds to the display requirements of different shifts or different operating conditions. Operators can switch between corresponding schemes with a single click using preset shortcut keys. The switching process directly replaces the configuration parameters in the shared memory without reloading the interface. The module supports importing and exporting configuration parameters and enables rapid migration of configurations between multiple sites through encrypted mobile storage devices. The migration process automatically performs compatibility checks on the target site to avoid display anomalies caused by mismatches between configuration parameters and the hardware environment. All configuration modification operations automatically generate operation logs, recording the modifier, modification time, and modification content for easy traceability and verification later.
[0012] Furthermore, the system has a built-in alarm linkage and positioning unit. Based on a globally unified measurement point label name, a mapping data table is established. The table pre-stores the monitoring screen number, screen coordinates, logical configuration page number, function block location, and historical trend storage address information corresponding to each measurement point label. Any alarm item in the alarm window can be left-clicked to jump to the corresponding monitoring screen. After jumping, the screen is automatically positioned in the coordinate area of the measurement point, and the corresponding measurement point is highlighted with a high-brightness border. Alarm items can be right-clicked to select the positioning configuration function, which directly jumps to the corresponding function block location in the logical configuration page. The screen is automatically scaled to fit the display ratio of the function block. Alarm items can be right-clicked to select the trend retrieval function, which directly pops up the real-time and historical trend window of the corresponding measurement point. The trend window displays the data curve for 10 minutes before and after the alarm is triggered by default. Abnormal data segments in the trend curve are associated with the corresponding alarm events by matching the measurement point label with the timestamp. Clicking on the abnormal segment will display the associated alarm details and handling records.
[0013] Furthermore, the intelligent voice alarm processing module supports zoned broadcast control. The system is divided into four zones according to the desulfurization process area: absorption tower zone, pulping zone, flue gas zone, and common zone. The zone division matches the work station responsibilities of on-site operators. Each operator station can be bound to one or more process zones. After binding, only the alarm signals of the bound zone are broadcast. Alarms of unbound zones are only displayed in the alarm window without triggering voice broadcast. The zone binding parameters can be adjusted online with administrator privileges and take effect in real time. The system supports a global broadcast mode. When an emergency alarm is triggered, the zone settings are automatically overridden. All operator stations broadcast emergency alarm information synchronously. The alarm status of multiple operator stations is synchronized in real time. After any station completes the alarm confirmation and blocking operation, the operation result is synchronized to all associated stations in real time through the DCS industrial Ethernet. The alarm list status of all stations remains consistent, avoiding repetitive operations or information discrepancies among multiple positions.
[0014] Furthermore, the alarm data source acquisition module synchronously collects the operating status parameters of each DCS node, including CPU utilization, memory utilization, network bandwidth utilization, and critical process running status, all in percentage (%). The module has a built-in node anomaly score calculation unit to calculate the anomaly score of a single node. The formula is: ,in This represents the real-time CPU utilization rate, with a value ranging from 0 to 100. This represents the real-time memory usage rate, with a value ranging from 0 to 100. This represents network bandwidth utilization, with a value ranging from 0 to 100. This is a statistical value for the number of process anomalies. Each time a critical process exits abnormally, 20 points are added, with a maximum of 60 points. Critical processes are defined as core processes that directly participate in alarm data collection, voice synthesis, and image rendering. , , , The weighting coefficients are 0.4, 0.3, 0.2, and 0.1, respectively. The weighting coefficients are calibrated based on the statistical data of the impact of domestic DCS hardware failures. When the abnormal score is higher than 30, a general level alarm is triggered; when it is higher than 60, an important level alarm is triggered; and when it is higher than 80, an emergency level alarm is triggered. All node abnormal alarms are uniformly included in the broadcast priority queue and are subject to the same sorting and broadcasting rules as process alarms.
[0015] Compared with existing technologies, this desulfurization DCS intelligent voice alarm and customized status display system has the following advantages: I. This invention addresses the characteristics of long process interlocking links in desulfurization systems and the tendency for single-point faults to trigger a batch of secondary alarms. It employs a dynamic priority broadcasting algorithm with correlation suppression, which distinguishes between root cause alarms and secondary chain alarms based on the desulfurization process interlocking topology. It calculates broadcasting weights from multiple dimensions, including alarm level, process impact, unit load, and equipment operating status, thereby mitigating the interference of alarm storms on operators from the root cause. This ensures that core fault information is broadcast first, effectively shortening the response time for fault location and handling in the desulfurization system and improving the stability of the unit's environmental protection facilities.
[0016] Second, this invention adopts a runtime shared memory online configuration architecture, which allows for the customization and switching of alarm display styles and window layouts and multiple schemes without entering the engineer configuration mode. It adapts to the operating habits of personnel on different shifts. With the bidirectional linkage and positioning mechanism of alarm points, monitoring screens, logic configurations, and trend curves, it eliminates the need for manual cross-module search for measurement points, reduces the monitoring burden and operational complexity of operators, and improves the overall efficiency of alarm handling and system maintenance.
[0017] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description
[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.
[0019] Figure 1 A flowchart of a desulfurization DCS intelligent voice alarm and customized status display system; Figure 2 A flowchart of dynamic calculation of alarm broadcast priority with correlation suppression for a desulfurization DCS intelligent voice alarm and custom status display system; Figure 3 This is a flowchart illustrating the multi-dimensional linkage positioning operation of alarm items in a desulfurization DCS intelligent voice alarm and custom status display system. Detailed Implementation
[0020] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.
[0021] Example 1: Daily Operation Scenario of the Second Phase Desulfurization DCS Localization Retrofit at Qingdao Company This embodiment is applied to the daily operation monitoring scenario of the DCS localization retrofit project for the flue gas desulfurization unit of the second phase 300MW unit of Qingdao Company. The desulfurization system includes a dual-tower dual-circulation absorption tower, two sets of slurry preparation systems, a flue gas bypass system and a common auxiliary system. There are more than 3,200 monitoring and control points in total. The system is deployed on the localized DCS platform to realize intelligent voice alarm and personalized display monitoring of the entire process of the desulfurization system.
[0022] At the system hardware level, the process control station adopts a domestically produced Phytium architecture DPU, and the operator station runs a domestically produced Kylin operating system. Each operator station is connected to an external microphone and a 27-inch industrial display terminal. All devices communicate via a 100Mbps industrial Ethernet network. During the system power-on initialization phase, the alarm data source acquisition module completes the communication link handshake with the process control station, synchronously loads the list of all station measurement points and the equipment operating status label mapping table, and establishes a bidirectional data transmission channel with the intelligent voice alarm processing module and the alarm status custom display module. During daily operation, the acquisition module scans all process measurement points and equipment status signals at a fixed 50ms cycle, covering core parameters such as absorber level, slurry pH value, flue gas inlet and outlet SO2 concentration, slurry circulation pump operating status, and damper position. It also synchronously acquires CPU utilization, memory utilization, network bandwidth, and key process status data for each operator station and DPU node. The acquired raw signals are processed by hardware-level filtering and de-jittering before being pushed to the two downstream core functional modules. The complete workflow of the overall system data acquisition, processing, voice broadcasting, and display update is as follows: Figure 1 As shown.
[0023] During initialization, the intelligent voice alarm processing module loads a corpus of desulfurization terminology and completes the startup calibration of the offline speech synthesis engine. The environmental noise acquisition unit collects ambient noise values in the control room at 1-minute intervals via an external microphone connected to the operator station, dynamically adjusting the broadcast volume based on the noise level. The system baseline volume is set to 60% of the maximum volume. For every 10dB increase in ambient noise, the broadcast volume increases by 15% relative to the baseline value. The upper and lower limits for volume adjustment are set from 30% to 100%, based on the typical noise range of 40dB to 70dB in the control room, balancing audibility and operator comfort. The module defaults to a silent period from 22:00 to 6:00 the next day, during which only emergency alarms trigger voice broadcasts; other alarm levels only trigger pop-up displays, adapting to the low-interference requirements of night shifts. The default broadcast interval is set to 60 seconds, which operators can customize within a range of 30 to 300 seconds. This range matches the typical change cycle of desulfurization process parameters, avoiding information redundancy caused by frequent repetitive broadcasts.
[0024] The alarm status custom display module loads the default display scheme during initialization. The real-time rendering unit reads the configuration parameters in the shared memory area of the running state at a fixed period of 100ms and refreshes the alarm window interface. This refresh cycle matches the general refresh standard of industrial monitoring systems, balancing real-time display and CPU load control. Operators can directly adjust the alarm item color, window size, and docking position through the built-in configuration entry in the interface. After the parameters are submitted, the configuration verification unit performs value range and format validity checks. Once the verification is successful, the parameters are directly written to the shared memory and take effect immediately in the next rendering cycle. The entire process does not require switching to the engineer configuration mode, restarting the system, or reloading the project file. The operation process for alarm items to quickly jump to the monitoring screen, logic configuration page, and historical trend curve through the linkage positioning unit is as follows: Figure 3 As shown. During system operation, the node anomaly score calculation unit calculates the anomaly score of each hardware node every 5 minutes. When the anomaly score exceeds 30 points, a general-level system alarm is triggered and included in a unified alarm queue for broadcast and display, thereby realizing the self-monitoring of the DCS system's own operating status.
[0025] In summary, this embodiment, through the coordinated operation of various modules, fully realizes the intelligent voice broadcast and personalized display functions of desulfurization system alarms on a domestically produced DCS hardware and software platform, adapting to the monitoring needs of power plant daily operation and effectively improving the ease of use and operational reliability of the alarm system.
[0026] Example 2: Alarm Storm Handling Scenario for Desulfurization Absorber Slurry Circulation Pump Failure This embodiment is applied to the scenario of a sudden electrical fault tripping of the No. 1 slurry circulation pump of the No. 1 absorber in the Phase II desulfurization system of Qingdao Company. It is used to verify the actual operation effect of the dynamic priority algorithm with correlation suppression in the scenario of batch cascading alarms, and solve the problem of alarm storms interfering with the fault judgment of operators in traditional alarm systems.
[0027] When the No. 1 slurry circulation pump of the No. 1 absorber suddenly trips due to an electrical circuit fault, the alarm data source acquisition module collects the pump's operating status change signal within the first scan cycle. This simultaneously triggers a total of 12 cascading alarm signals, including an alarm for the pump's outlet damper closing, a low slurry circulation flow alarm in the absorber, an alarm for fluctuations in the absorber level, and an alarm for rising SO2 concentration at the outlet. This creates a storm of concurrent alarms. The dynamic weighted calculation process for alarm broadcast priority with correlation suppression is as follows: Figure 2 As shown, the priority calculation unit calculates the broadcast priority score P for each triggered alarm, distinguishing between root cause alarms and secondary alarms based on the pre-stored desulfurization process interlocking topology.
[0028] The tripping of the No. 1 slurry circulation pump is an independently triggered root cause fault, with an alarm level of emergency (G = 100 points). This measuring point belongs to the core subsystem of the absorption tower, and the calibration value T is taken as 1.0. The current unit load is 90% of the rated load, so L is taken as 1.2. The equipment is in operation, so S is taken as 1.0. The correlation suppression coefficient for the root cause alarm... Set to 1.0; weighting coefficient , , Taking values of 0.6, 0.2, and 0.2 respectively, and substituting them into the formula, we get... =(0.6×100+0.2×1.0+0.2×1.2)×1.0×1.0=60.44 points. Of the remaining 11 secondary interlocking alarms, the exit damper closure alarm is classified as critical. 70 points; Take 0.8; Take 1.2; Set to 1.0; this alarm is a direct secondary alarm caused by pump tripping, and is classified as a first-level secondary alarm. Take 0.6; calculate to get =(0.6×70+0.2×0.8+0.2×1.2)×1.0×0.6=25.44 points. Other secondary alarms include low flow rate and level fluctuations. The value is set to 0.4-0.5, and the calculated priority scores are all significantly lower than those of the root cause alarm. The value of the correlation suppression coefficient is set according to the information reception saturation of desulfurization operation and maintenance personnel. After the secondary alarm is downweighted, it ensures that fault information is not missed, but also does not preempt the broadcasting resources of the core fault.
[0029] All alarms press Values are sorted from highest to lowest, with root cause alarms at the top of the broadcast queue, followed by secondary alarms. During broadcast, the root cause of the pump trip is broadcast first, followed by secondary alarms in priority order to avoid information confusion caused by multiple alarms broadcasting simultaneously. For this interlocking protection action, the system automatically classifies it as an emergency level, triggering a full-screen pop-up prompt. The pop-up displays the fault name, action time, and a list of associated equipment, which must be manually confirmed and closed by the operator. At the same time, the condition-adaptive alarm suppression unit automatically identifies the No. 2 slurry circulation pump in standby mode and blocks the corresponding idle measuring point alarms to prevent invalid information from interfering with the operator's judgment.
[0030] In summary, this embodiment uses a dynamic priority algorithm with correlation suppression to prioritize the broadcast of root cause faults in batch cascading alarm scenarios, effectively mitigating the interference of alarm storms on operators, shortening fault location response time, and improving the fault handling efficiency of the desulfurization system.
[0031] Example 3: Customized display application scenario for four-value, three-shift switching This embodiment is applied to the shift handover scenario of the four-shift, three-operation mode of the desulfurization system in Phase II of Qingdao Company. The operating habits and monitoring focuses of operators on different shifts are different. This is used to verify the actual application effect of the custom configuration of alarm display and the function of quick switching between multiple schemes.
[0032] The second phase of desulfurization operation at Qingdao Power Plant adopts a four-shift, three-rotation scheduling model. The four operating shifts rotate between day, afternoon, and night shifts. The monitoring habits of operators on different shifts differ; day shift personnel focus on displaying all alarm details, while night shift personnel emphasize prominent notifications for high-level alarms. The alarm status custom display module has eight built-in independent display schemes: four corresponding to the day shift configurations for the four operating shifts, three to the night shift configurations, and one as an emergency backup scheme. The number of schemes matches the actual scheduling needs of the power plant, avoiding wasted storage space and scheme redundancy. Each scheme is pre-configured with different alarm color indicators, window sizes, number of items per page, and flashing frequency. The day shift scheme displays 20 alarms per page with moderate font size, arranged in reverse chronological order; the night shift scheme displays 10 alarms per page, with high-level alarms displayed in large, bright, flashing font, and low-level alarms displayed less prominently.
[0033] During shift handover, incoming operators can switch to the corresponding value display scheme with a single click using preset function shortcut keys. The switching command directly modifies the configuration parameters in the shared memory of the operating state. The real-time rendering unit updates the interface within the next 100ms refresh cycle, ensuring a smooth switching process without screen lag or affecting the real-time acquisition and display of alarm data, and without needing to exit the operation monitoring mode. If operators need to fine-tune the display parameters to suit their personal preferences, they can directly adjust the background color and window docking position of alarm entries through the online configuration interface. After adjustment, clicking "confirm" will take effect. The configuration verification unit automatically verifies the color parameter format and window size boundaries to avoid display anomalies caused by illegal parameters. All modification operations automatically generate operation logs, recording the operator's account, modification time, and parameters before and after the modification, meeting the power plant's operation and maintenance traceability management requirements.
[0034] Configuration schemes support importing and exporting via encrypted mobile storage devices. When adding a new operator station, a mature configuration scheme can be directly imported, eliminating the need for point-to-point reconfiguration and shortening the engineering debugging cycle. During migration, the system automatically verifies the target station's screen resolution and system version, automatically adapting display parameters to avoid interface misalignment issues. Simultaneously, with the zoned broadcast function, each operating station is bound to its corresponding process zone; the absorber station is bound to the absorber zone, the pulping station to the pulping zone, and alarm voices are broadcast only for the responsible area, reducing information interference from cross-zone alarms. Emergency alarms automatically trigger global broadcasts, with all stations broadcasting simultaneously to ensure no major faults are missed.
[0035] In summary, this embodiment, through the storage of multiple display schemes and the one-click switching function, adapts to the operating habits of operators on different shifts. The online configuration and real-time effective architecture do not affect normal operation and monitoring, effectively improving the personnel adaptability and operation and maintenance flexibility of the alarm system.
[0036] Example 4: System Performance and Effect Data Verification Example This embodiment is applied to the factory performance test scenario after the second phase desulfurization DCS upgrade of Qingdao Company. It is used to verify the functional effectiveness and performance indicators of this system, compare the actual operation effect of traditional alarm systems, and verify the engineering practicality of the solution.
[0037] The tests were conducted on a domestically produced DCS platform. The test environment was configured with Phytium FT2000 / 4DPU, Kylin V10 operating system, and 100Mbps industrial Ethernet, connecting 3200 simulated test points. Tests were conducted simulating both stable daily operation and typical fault conditions. In the performance test, when a single station was fully loaded with 3200 test points, the refresh latency from alarm data acquisition by the DPU to display on the operator station was 85ms, meeting the conventional requirement of no more than 100ms for industrial monitoring systems; the voice synthesis response time was 120ms, and the total latency from alarm triggering to voice broadcast initiation was less than 200ms, meeting the response requirements for real-time alarms; the time for the interface to take effect after display configuration changes was 110ms, achieving seamless switching without interfering with operation monitoring.
[0038] In the functional effectiveness test, 10 typical desulfurization equipment failures were simulated, with each failure triggering an average of 15 cascading alarms. Traditional alarm systems broadcast alarms sequentially based on trigger time, requiring operators an average of 18 seconds to locate the root cause of the fault. After adopting the correlation suppression priority algorithm of this system, the average time for operators to locate the root cause of the fault was reduced to 7 seconds, improving fault location efficiency by approximately 61%. In the false alarm rate test, for measuring points with large fluctuations in liquid level and pressure, enabling the corresponding delay trigger function reduced the number of false alarms from an average of 12 per day to an average of 2 per day, a reduction of approximately 83%, effectively reducing the interference of invalid alarms on operators.
[0039] In configuration efficiency testing, traditional systems require entering engineer configuration mode, modifying, and then reinstalling the project, with each adjustment taking an average of 12 minutes and necessitating the pausing of some monitoring functions. Using the online configuration function of this system, the average adjustment time is reduced to 40 seconds, with no interruption to monitoring throughout the process, resulting in a significant improvement in configuration efficiency. In node self-monitoring function testing, simulating two scenarios—DPU node CPU utilization rising to 85% and critical processes abnormally exiting—the system triggered corresponding system alarms within one scan cycle in both cases. The alarm information was accurate, assisting maintenance personnel in quickly identifying potential system faults.
[0040] In summary, this embodiment verifies that the system outperforms traditional alarm systems in terms of real-time performance, fault location efficiency, false alarm suppression, and configuration flexibility. All performance indicators meet the industrial operation requirements of desulfurization DCS, demonstrating good engineering application value and promising prospects for promotion.
[0041] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.
Claims
1. A desulfurization DCS intelligent voice alarm and customized status display system, characterized in that, The system includes the following components: Alarm data source acquisition module: It connects to the desulfurization DCS process control station and the field measurement and control equipment. The output end is connected to the input end of the intelligent voice alarm processing module and the alarm status custom display module respectively. The alarm data source acquisition module synchronously collects the desulfurization process measurement point over-limit signals, equipment displacement signals, protection action signals and the corresponding equipment operation status label data. Intelligent voice alarm processing module: It has a built-in priority calculation unit and an offline voice synthesis unit. It dynamically calculates the broadcast priority of each alarm based on the alarm level, the degree of process impact, the equipment operating status, and the unit load parameters, and executes the corresponding mode of voice broadcast according to the priority. Alarm status custom display module: It has a built-in online configuration unit and real-time rendering unit. It receives the display parameters input by the operator and performs legality verification. The parameters that pass the verification are written into the running state cache and take effect directly.
2. The desulfurization DCS intelligent voice alarm and customized status display system according to claim 1, characterized in that, The intelligent voice alarm processing module's offline voice synthesis unit has a built-in corpus of desulfurization terminology. The corpus covers professional terms and alarm descriptions for the entire process of desulfurization tower, slurry circulation pump, flue gas heat exchanger, absorbent preparation, and wastewater treatment. The module has a built-in environmental noise acquisition unit that collects the ambient noise level in real time through an external pickup device connected to the operator station. The volume of the voice broadcast is dynamically adjusted according to the noise level. For every 10dB increase in noise, the broadcast volume is increased by 15% relative to the system's baseline volume. The volume adjustment range is 30%-100%. The module supports setting a silent period. During the silent period, only the highest level alarm is broadcast, while other alarm levels are only triggered to display a prompt. The module supports custom broadcast intervals.
3. The desulfurization DCS intelligent voice alarm and customized status display system according to claim 1, characterized in that, The priority calculation unit uses a dynamic weighted algorithm with correlation suppression to calculate the broadcast priority score of a single alarm. The formula is: ,in The base score for alarm level, This is the process influence coefficient. This is the unit load correction factor. For equipment operating status coefficient, This is the alarm correlation suppression coefficient. , , These are the weighting coefficients.
4. The desulfurization DCS intelligent voice alarm and customized status display system according to claim 1, characterized in that, The intelligent voice alarm processing module has a built-in adaptive alarm suppression unit with built-in equipment operation status association rules. It judges the equipment status by collecting equipment hard-contact operation feedback signals from the process control station. When the equipment is in a shutdown, maintenance, or isolation state, it automatically blocks all alarms corresponding to that equipment. The blocking state switches in real time according to the equipment operation feedback signal. The module supports setting alarm delay trigger function. For process measuring points with different fluctuation characteristics, a trigger delay of 0.5s-10s can be set. For measuring points with rapid fluctuations such as liquid level and pressure, a delay of 0.5s-2s is set. For measuring points with gradual changes such as pH and concentration, a delay of 3s-10s is set. The formal alarm is triggered only after the measuring point exceeds the delay threshold. For alarms of desulfurization system interlock protection action, the module automatically classifies them as emergency level, and the voice broadcast is immediately inserted at the top of the queue, while triggering a full-screen pop-up prompt.
5. The desulfurization DCS intelligent voice alarm and customized status display system according to claim 1, characterized in that, The online configuration unit of the alarm status custom display module provides a visual interactive interface. Operators can directly adjust the alarm window parameters through the operator station interface. After the parameters are input, the configuration verification unit performs a legality verification. Parameters that pass the verification are directly written to the running shared memory area. The real-time rendering unit reads the configuration parameters in the shared memory at a fixed period of 100ms and refreshes the alarm display interface. All monitoring interfaces of the same operator station share the same shared memory area, and the configuration parameters are automatically synchronized to all open alarm windows and monitoring screens.
6. The desulfurization DCS intelligent voice alarm and customized status display system according to claim 1, characterized in that, The alarm status custom display module has a built-in configuration scheme storage unit, which supports saving up to 8 independent display schemes. The number of schemes matches the power plant's four-shift, three-operation shift configuration plus two emergency backup schemes. Each scheme corresponds to the display requirements of different shifts or different operating conditions. Operators can switch the corresponding scheme with one click using preset shortcut keys. The module supports the import and export of configuration parameters and completes rapid configuration migration between multiple sites through encrypted mobile storage devices. The migration process automatically completes the compatibility verification of the target site, and all configuration modification operations automatically generate operation logs.
7. The desulfurization DCS intelligent voice alarm and customized status display system according to claim 1, characterized in that, The system has a built-in alarm linkage and positioning unit. Based on a globally unified measurement point label name, a mapping data table is established. Any alarm entry in the alarm window can be left-clicked to jump to the corresponding monitoring screen. After jumping, the screen automatically positions itself in the coordinate area of the measurement point, and the corresponding measurement point is highlighted with a high-brightness border. Alarm entries can be right-clicked to select the positioning configuration function, which directly jumps to the corresponding function block in the logic configuration page. The screen automatically scales to fit the display ratio of the function block. Alarm entries can also be right-clicked to select the trend retrieval function, which directly pops up the real-time and historical trend windows of the corresponding measurement point. The trend window displays the data curves for 10 minutes before and after the alarm is triggered by default. Abnormal data segments in the trend curve are associated with the corresponding alarm events by matching the measurement point label with the timestamp. Clicking on the abnormal segment will display the associated alarm details and handling records.
8. The desulfurization DCS intelligent voice alarm and customized status display system according to claim 1, characterized in that, The intelligent voice alarm processing module supports zoned broadcast control. The system is divided into four zones according to the desulfurization process area: absorption tower zone, pulping zone, flue gas zone, and common zone. The zone division matches the work station responsibilities of the on-site operators. Each operator station can be bound to one or more process zones. After binding, only the alarm signals of the bound zone are broadcast. Alarms of unbound zones are only displayed in the alarm window and do not trigger voice broadcast. The zone binding parameters can be adjusted online with administrator privileges and take effect in real time. The system supports global broadcast mode. When an emergency alarm is triggered, the zone settings are automatically overridden. All operator stations broadcast emergency alarm information synchronously. The alarm status of multiple operator stations is synchronized in real time. After any station completes the alarm confirmation and blocking operation, the operation result is synchronized to all associated stations in real time through the DCS industrial Ethernet.
9. The desulfurization DCS intelligent voice alarm and customized status display system according to claim 1, characterized in that, The alarm data source acquisition module synchronously collects the operating status parameters of each node in the DCS. The module has a built-in node anomaly score calculation unit to calculate the anomaly score of a single node. The formula is: ,in CPU real-time utilization Real-time memory usage Network bandwidth utilization This is a statistical value for the number of process exceptions. , , , These are the weighting coefficients.