Information-based intelligent diagnosis and fault early warning method for operation state of port environmental protection equipment

Abstract

The application discloses a port environmental protection equipment operation state information intelligent diagnosis and fault early warning method, relates to the technical field of port environmental protection equipment intelligent operation and maintenance and predictive maintenance, and comprises the following steps: a central analysis terminal analyzes and preprocesses time series data flow to obtain a structure health coefficient, an energy consumption deviation coefficient and an environmental corrosion rate coefficient; the structure health coefficient, the energy consumption deviation coefficient and the environmental corrosion rate coefficient are input into an equipment degradation comprehensive model; and an equipment degradation comprehensive index is output. The application generates a dynamic early warning threshold value through an early warning decision terminal, divides a first, second and third early warning in combination with the equipment degradation comprehensive index and a moving standard deviation, replaces a traditional fixed threshold value determination mode, makes the early warning dynamically adapt to equipment operation conditions and environmental changes, and significantly improves the accuracy and timeliness of fault early warning.

Description

Technical Field

[0001] This invention relates to the field of intelligent operation and maintenance and predictive maintenance technology for port environmental protection equipment, and in particular to an information-based intelligent diagnosis and fault early warning method for the operating status of port environmental protection equipment. Background Technology

[0002] With the acceleration of global trade integration and the continuous upgrading of green port construction standards, the stable operation of environmental protection equipment in ports, such as dust collectors, sewage treatment pumps, shore power frequency converters, and seawater desalination systems, is directly related to the effectiveness of ecological environment governance and operational economics, as ports serve as important logistics hubs and energy consumption nodes. Especially in the special environment of high salt spray and high humidity along the coast, port environmental protection equipment is subjected to multiple stresses such as mechanical vibration, energy consumption fluctuations, and environmental corrosion for a long time, and its performance degradation and failure risk are significantly higher than those of ordinary industrial equipment.

[0003] Currently, the condition monitoring and fault diagnosis technology for port environmental protection equipment is still dominated by traditional methods. On the one hand, it relies on manual inspections and periodic shutdowns for testing, using tools such as portable vibration meters and multimeters to collect discrete data, and combining this with the experience of maintenance personnel to judge the health status of the equipment. This method has limitations such as long data collection intervals, limited coverage, and significant influence from subjective experience. On the other hand, although some ports have introduced single-dimensional online monitoring systems, such as simply recording vibration or energy consumption data and issuing alarms for exceeding limits, they lack the ability to collaboratively analyze multi-source heterogeneous data such as mechanical structure, energy efficiency, and environmental corrosion, making it difficult to build a comprehensive assessment model for equipment deterioration. At the same time, existing technologies mostly use fixed thresholds for fault determination, without considering the individual differences of equipment, dynamic changes in operating conditions, and the coupled effects of environmental factors, resulting in insufficient accuracy and timeliness of early warnings.

[0004] The core problems facing current technological development are concentrated in three dimensions: First, the ability to process and integrate multi-source data is weak. Due to the large differences in the collection frequency, dimensions, and characteristics of data such as vibration spectrum, energy efficiency flow, and meteorological environmental corrosion, traditional methods are difficult to achieve effective synchronization and correlation analysis, resulting in one-sided equipment condition assessment. Second, the fault diagnosis and early warning mechanisms are lagging behind. They rely heavily on historical fault data for post-event analysis and lack the ability to predict equipment degradation trends based on real-time data, making it difficult to achieve the transformation from fault repair to preventive maintenance. Third, intelligent response and operation and maintenance linkage are insufficient. Existing systems mostly remain at the level of alarm prompts and fail to dynamically adjust monitoring strategies according to the warning level. They also cannot automatically generate appropriate maintenance work orders and link with the port equipment management system to perform operations such as spare parts allocation and shutdown isolation, resulting in low efficiency in fault handling and easy to cause secondary losses.

[0005] Therefore, it is essential to develop an information-based intelligent diagnostic and fault early warning method for the operating status of port environmental protection equipment to solve the above problems. Summary of the Invention

[0006] The purpose of this invention is to provide an information-based intelligent diagnosis and fault early warning method for the operating status of port environmental protection equipment, so as to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: an information-based intelligent diagnosis and fault early warning method for the operating status of port environmental protection equipment, comprising an environmental protection equipment data acquisition terminal, an edge computing terminal, a dynamic monitoring terminal, a central analysis terminal, an early warning decision terminal, and an operation and maintenance execution terminal, specifically including the following steps:

[0008] S1. The environmental protection equipment data acquisition terminal collects multi-dimensional operational data from the port environmental protection equipment itself and its operating environment through a multi-source sensor network, forming a raw data set of equipment status; the raw data set of equipment status includes three-dimensional vibration spectrum data, energy efficiency flow data, and meteorological environmental corrosion data.

[0009] S2. The edge computing terminal uses the Fourier transform algorithm to denoise the three-dimensional vibration spectrum data, the Kalman filter algorithm to dynamically calibrate the energy efficiency stream data, and the time series alignment algorithm to synchronize the meteorological and environmental corrosion data, generating a preprocessed time series data stream.

[0010] S3. The central analysis terminal analyzes and preprocesses the time-series data stream to obtain the structural health coefficient, energy consumption deviation coefficient, and environmental corrosion rate coefficient. It then inputs the structural health coefficient, energy consumption deviation coefficient, and environmental corrosion rate coefficient into the equipment deterioration comprehensive model and outputs the equipment deterioration comprehensive index.

[0011] S4. The early warning decision terminal generates a dynamic early warning threshold based on the comprehensive equipment deterioration index. When the comprehensive equipment deterioration index exceeds the dynamic early warning threshold, a hierarchical early warning instruction set is triggered.

[0012] S5. The dynamic monitoring terminal parses the hierarchical early warning instruction set and dynamically adjusts the sampling frequency and monitoring dimension parameters of the multi-source sensor network according to the early warning level.

[0013] S6, the operation and maintenance execution terminal generates equipment maintenance work orders based on the hierarchical early warning instruction set, and links port equipment to perform shutdown isolation, spare parts allocation and performance optimization operations.

[0014] Preferably, the three-dimensional vibration spectrum data includes the maximum axial amplitude deviation, the radial vibration frequency offset, and the standard deviation of torque fluctuation; the energy efficiency stream data includes the unit energy consumption output power, the proportion of reactive power, and the motor efficiency decay slope; and the meteorological environment corrosion data includes the instantaneous value of salt spray concentration, the amplitude of humidity stress fluctuation, and the cumulative amount of temperature difference cycle.

[0015] Preferably, the structural health coefficient is specifically:

[0016] ,

[0017] Where α is the amplitude tolerance factor, β is the frequency sensitivity factor, λ is the torque fluctuation gain factor, and δ max σ is the maximum axial amplitude deviation. v Let be the standard deviation of torque fluctuation, Δf be the radial vibration frequency offset, ln be the logarithm to the base e, and e be the natural constant.

[0018] Preferably, the energy consumption deviation coefficient is specifically:

[0019] ,

[0020] Among them, P e P0 is the power output per unit energy consumption, P0 is the rated operating condition reference power, and η is the power output per unit energy consumption. rea η represents the proportion of reactive power. ref k is the reactive power threshold. e Let be the slope of the motor efficiency decay, tan be the tangent function, and π be pi.

[0021] Preferably, the environmental corrosion rate coefficient is specifically:

[0022] ,

[0023] Among them, C s C0 is the instantaneous value of salt spray concentration, and A is the safe threshold for salt spray concentration. h A0 represents the humidity stress fluctuation amplitude, and T represents the permissible humidity stress amplitude. q T0 is the cumulative temperature difference cycle, T0 is the cumulative temperature difference cycle threshold, and min is the minimum value function.

[0024] Preferably, the comprehensive equipment degradation model is as follows:

[0025] ,

[0026] The calculation rules for the dynamic weights ω1, ω2, and ω3 are as follows:

[0027] ,

[0028] ,

[0029] ,

[0030] Where I is the comprehensive equipment deterioration index, S is the structural health coefficient, D is the energy consumption deviation coefficient, R is the environmental corrosion rate coefficient, Δt is the continuous monitoring time interval, τ is the time decay constant, e is the natural constant, D(t) is the energy consumption deviation coefficient value at the current time t, D(t-Δt) is the energy consumption deviation coefficient value at the previous monitoring time t-Δt, R(t) is the environmental corrosion rate coefficient value at the current time t, and R(t-Δt) is the environmental corrosion rate coefficient value at the previous monitoring time t-Δt.

[0031] Preferably, the mechanism for generating the dynamic early warning threshold is as follows:

[0032] ,

[0033] Where T is the dynamic warning threshold, δ is the sensitivity coefficient, and μ I σ is the moving average of the comprehensive equipment deterioration index. I γ is the moving standard deviation of the equipment degradation composite index, and γ is the baseline adjustment factor.

[0034] Preferably, the tiered early warning instruction set includes:

[0035] Emergency shutdown judgment command is recorded as Level 1 warning, critical maintenance command is recorded as Level 2 warning, and performance degradation warning command is recorded as Level 3 warning;

[0036] The classification rules for the Level 1 warning are as follows: ;

[0037] The classification rules for the Level II early warning are as follows: ;

[0038] The classification rules for the three-level early warning are as follows: ;

[0039] Where, σ I Let I be the moving standard deviation of the comprehensive equipment deterioration index, T be the dynamic early warning threshold, and I be the comprehensive equipment deterioration index.

[0040] Preferably, the dynamic adjustment in step S5 includes:

[0041] When the warning level is 1, the sampling frequency of the vibration spectrum data is increased to 3 times the baseline value, and the monitoring dimension of the temperature sensor is added.

[0042] When the warning level is Level II, the sampling frequency of energy efficiency stream data will be increased to twice the baseline value, and the real-time calibration mode for corrosion data will be activated.

[0043] When the warning level is 3, the default sampling frequency is maintained, and the monitoring weight of the axial amplitude deviation of the spectrum data is increased.

[0044] Preferably, the operation in step S6 includes:

[0045] When a Level 1 warning is issued, an emergency shutdown work order is generated, triggering the equipment's power isolation protocol.

[0046] When the warning level is level 2, a preventive maintenance work order is generated, and spare motors and desalination consumables are automatically allocated.

[0047] When a Level 3 warning is issued, a performance calibration work order is generated to adjust the reactive power compensation parameters of the equipment.

[0048] The technical effects and advantages of this invention are as follows:

[0049] 1. This invention calculates the structural health coefficient, energy consumption deviation coefficient, and environmental corrosion rate coefficient through a central analysis terminal, and outputs a comprehensive equipment deterioration index based on a dynamic weighted equipment deterioration comprehensive model. This achieves a quantitative and integrated assessment of equipment mechanical structure, energy efficiency, and environmental corrosion, and improves the comprehensiveness and accuracy of equipment health status diagnosis.

[0050] 2. This invention generates dynamic early warning thresholds through an early warning decision terminal, and divides the warnings into first-level, second-level, and third-level levels by combining the comprehensive equipment deterioration index and the moving standard deviation. This replaces the traditional fixed threshold judgment mode, enabling the early warning to dynamically adapt to changes in equipment operating conditions and environment, and significantly improving the accuracy and timeliness of fault early warning.

[0051] 3. This invention adjusts the sampling frequency and monitoring dimensions of a multi-source sensor network according to the warning level through a dynamic monitoring terminal. For example, a first-level warning increases the vibration sampling frequency and adds temperature monitoring, while a second-level warning increases the energy data sampling frequency. This achieves dynamic optimization of monitoring resources and enhances the ability to perceive different fault risks.

[0052] 4. This invention generates corresponding maintenance work orders based on hierarchical early warning through the operation and maintenance execution terminal, and performs emergency shutdown, spare parts allocation and parameter optimization operations in a coordinated manner, thus constructing a closed-loop operation and maintenance system from early warning to handling, which greatly improves the efficiency of fault response and the timeliness of equipment maintenance. Attached Figure Description

[0053] Figure 1 This is a schematic diagram of the device connection according to the present invention.

[0054] Figure 2 This is a schematic diagram of the method steps of the present invention. Detailed Implementation

[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] This invention provides, for example Figure 1 The equipment connection diagram shown includes an environmental protection equipment data acquisition terminal, an edge computing terminal, a dynamic monitoring terminal, a central analysis terminal, an early warning decision-making terminal, and an operation and maintenance execution terminal;

[0057] This invention provides, for example Figure 2 The information-based intelligent diagnosis and fault early warning method for the operating status of port environmental protection equipment shown includes the following steps:

[0058] S1. The environmental protection equipment data acquisition terminal collects multi-dimensional operational data from the port environmental protection equipment itself and its operating environment through a multi-source sensor network, forming a raw data set of equipment status; the raw data set of equipment status includes three-dimensional vibration spectrum data, energy efficiency flow data, and meteorological environmental corrosion data.

[0059] Furthermore, in the above technical solution, the three-dimensional vibration spectrum data includes the maximum axial amplitude deviation, the radial vibration frequency offset, and the standard deviation of torque fluctuation; the energy efficiency stream data includes the unit energy consumption output power, the proportion of reactive power, and the motor efficiency decay slope; the meteorological environment corrosion data includes the instantaneous value of salt spray concentration, the amplitude of humidity stress fluctuation, and the cumulative amount of temperature difference cycle.

[0060] It should be noted that the maximum axial amplitude deviation is obtained by real-time collection of vibration acceleration data by a triaxial accelerometer installed at key axial positions of the equipment's transmission shaft, such as the motor bearing housing and the gearbox output shaft. The data is then converted into displacement amplitude through integration. The environmental protection equipment data acquisition terminal continuously tracks the peak amplitude during the continuous operation cycle and finally extracts the maximum deviation value as the maximum axial amplitude deviation. This value directly reflects the degree of axial imbalance of the mechanical structure.

[0061] The radial vibration frequency offset is calculated by capturing the vibration time-domain signal of the rotor or bearing of the equipment using a radial acceleration sensor and a vibration spectrum analyzer. This signal is then converted into a frequency domain spectrum using a fast Fourier transform. The environmental protection equipment data acquisition terminal automatically compares the current characteristic frequencies, such as the bearing pass frequency and gear meshing frequency, with the reference frequency under rated operating conditions to calculate the absolute offset Δf = |f|. 当前 -f 基准 This value quantifies the abnormal dynamic characteristics of rotating parts caused by wear or loosening;

[0062] The torque fluctuation standard deviation is obtained by monitoring the instantaneous torque value of the drive shaft, such as the coupling or power transmission shaft, using a non-contact torque sensor or strain gauge to generate continuous time series data; statistical analysis is performed on the torque fluctuation within a preset time window, such as 10 seconds, and the torque fluctuation standard deviation is calculated. This value reveals the torque instability caused by sudden changes in equipment load or damage to the transmission system.

[0063] The unit energy consumption output power is monitored in real time by smart meters deployed on port environmental protection equipment to monitor the total active power input. Simultaneously, the effective output power is accurately measured using mechanical sensors matched to the equipment type: for fans or pumps, flow meters and pressure sensors are linked to obtain flow and pressure difference data, and the real-time mechanical power is calculated using the formula: Output Power = Flow Rate × Pressure Difference × Medium Density Coefficient; for conveying equipment, the output power is calculated using a weighing sensor and a speed sensor: Output Power = Load Weight × Conveying Speed ​​× Friction Coefficient; based on this, the formula... Calculated;

[0064] The reactive power ratio is calculated using an embedded power quality analysis chip, such as the ADE9000, which acquires three-phase voltage and current waveforms in real time. The fundamental and harmonic components are decomposed using a Fast Fourier Transform to directly calculate the instantaneous reactive power and apparent power, which are then calculated according to the formula... Obtain the reactive power ratio;

[0065] The motor efficiency decay slope is obtained through The instantaneous electromechanical conversion efficiency is calculated, where the input electrical power is synchronously provided by a smart meter, and the mechanical output power is measured by a non-contact torque sensor, such as the HBMT40B, in conjunction with a photoelectric encoder. The edge computing terminal uses Kalman filtering to reduce noise in the instantaneous efficiency value, eliminating short-term fluctuations. The central analysis terminal performs linear regression analysis on a smoothed efficiency dataset spanning 72 consecutive hours to extract the slope parameter k. e ;

[0066] The instantaneous value of the salt spray concentration is collected in real time by an electrochemical salt spray sensor, such as the Alphasense SO2-B4 series, deployed within 1 meter of the windward side of the equipment body.

[0067] The humidity stress fluctuation amplitude relies on a high-precision temperature and humidity sensor array, such as the SHT85 sensor, deployed close to the outer surface of the equipment's metal structure, such as the motor housing or control cabinet. The sensor continuously captures the relative humidity time series data of the equipment surface at a sampling frequency of ≥1Hz, and performs statistical analysis on the humidity sequence within a specific time window, such as 10 minutes, to calculate its moving standard deviation as the humidity stress fluctuation amplitude.

[0068] The accumulated temperature difference is achieved by deploying a high-precision temperature and humidity sensor array, such as the SHT85 model, on the outer surface of the metal structure of port equipment, such as motor housings and control cabinets. Its physical significance lies in quantifying the accumulated metal fatigue stress caused by repeated thermal expansion and contraction, directly related to the risk of material damage caused by the diurnal temperature difference in the port. During data collection, the sensors continuously monitor the surface temperature at a sampling frequency of ≥1Hz and record the instantaneous temperature value T at adjacent time points in real time. i and T i-1 Then calculate the single temperature difference change ΔT i =|T i -T i-1 Finally, within a preset time window, such as within 24 hours, the effective △T i Perform weighted cumulative calculation: Where N is the effective temperature difference event, t i The weighting coefficient is 1.2 for periods lasting more than 10 minutes and 1.0 for short-term fluctuations.

[0069] S2. The edge computing terminal uses the Fourier transform algorithm to denoise the three-dimensional vibration spectrum data, the Kalman filter algorithm to dynamically calibrate the energy efficiency stream data, and the time series alignment algorithm to synchronize the meteorological and environmental corrosion data, generating a preprocessed time series data stream.

[0070] S3. The central analysis terminal analyzes and preprocesses the time-series data stream to obtain the structural health coefficient, energy consumption deviation coefficient, and environmental corrosion rate coefficient. It then inputs the structural health coefficient, energy consumption deviation coefficient, and environmental corrosion rate coefficient into the equipment deterioration comprehensive model and outputs the equipment deterioration comprehensive index.

[0071] Furthermore, in the above technical solution, the structural health coefficient specifically refers to:

[0072] ,

[0073] Where α is the amplitude tolerance factor, β is the frequency sensitivity factor, λ is the torque fluctuation gain factor, and δ max σ is the maximum axial amplitude deviation. v Let be the standard deviation of torque fluctuation, Δf be the radial vibration frequency offset, ln be the logarithm to the base e, and e be the natural constant.

[0074] It is important to know that the amplitude tolerance factor α is used to quantify the equipment's tolerance to axial vibration, and its physical basis lies in the material strength and stiffness characteristics of the equipment structure; when setting it, the axial vibration safety threshold δ provided by the equipment manufacturer must be obtained. safe Based on historical normal operation data, the maximum axial amplitude deviation δ was statistically calculated. max 95th percentile δ 95, where the axial vibration safety threshold is δ safe Ensure that the 95th percentile value δ based on historical data statistics 95 does not exceed δ safe , and calculate the initial value according to the formula , ensure that when δ max ≤δ 95 , the Gaussian term in the structural health factor S that is, tolerate a 5% deviation probability;

[0075] The frequency sensitivity factor β is used to characterize the risk weight of the radial vibration frequency offset △f to the equipment structure, and its core is to avoid resonance effects; determine the natural frequency f of the equipment through modal analysis n After that, select the gain coefficient k and calculate the initial value according to the formula , where f0 is the radial vibration frequency under rated conditions, and the value range of the gain coefficient k is 0.1~0.5, and the default setting is 0.3, which can be adjusted. For example, in the high salt fog environment of coastal ports, the motor is easily corroded, resulting in mass imbalance and increasing the risk of frequency offset. Increase k to 0.4;

[0076] The torque fluctuation gain factor λ is used to amplify the impact of the torque fluctuation standard deviation σ v on the structural health. The specific setting of its value is: first calculate the torque design margin M through the formula , and then set λ in different ranges according to the value of M: when M≤1.2, such as the transmission main shaft of an old dust removal fan, due to long-term corrosion, the fatigue limit of the material decreases. Take λ = 1.5 to amplify the risk of low margin. When 1.2 < M ≤ 1.5, such as the variable frequency drive system of a sewage treatment pump, it needs to adapt to load fluctuations but the design redundancy is moderate. Take λ = 1.2 to moderately increase the sensitivity. When M > 1.5, such as the voltage stabilizing gearbox of an onshore power frequency converter, the design redundancy is high and the operation is stable. Take λ = 1.0 to avoid false warnings; where, T design is the design limit torque, and the nominal value is directly obtained from the transmission component technical manual provided by the equipment manufacturer; T rated is the maximum working torque under rated conditions. For equipment equipped with a strain torque sensor, such as HBMT40B, continuously collect the operation data of the transmission shaft for 72 hours, and take the 99th percentile of the torque instantaneous value as T rated ;

[0077] Furthermore, in the above technical solution, the energy consumption deviation coefficient is specifically:

[0078] ,

[0079] where, P e is the unit energy consumption output power, P0 is the rated condition reference power, η rea is the reactive power ratio, η refk is the reactive power threshold. e Let be the slope of the motor efficiency decay, tan be the tangent function, and π be pi.

[0080] It should be noted that the rated operating condition reference power P0 is determined based on the equipment nameplate data or the manufacturer's technical manual, but needs to be dynamically adjusted according to real-time environmental parameters. For example, in a high-temperature and high-humidity port environment, the motor efficiency attenuation is compensated according to the IEC60034 standard: when the ambient temperature exceeds 40°C, P0 needs to be reduced by 2% for every 5°C increase.

[0081] The reactive power threshold η ref By analyzing historical operational data, η was taken from a continuous 90-day period. rea The 90th percentile value is used as the reactive power threshold.

[0082] Furthermore, in the above technical solution, the environmental corrosion rate coefficient is specifically:

[0083] ,

[0084] Among them, C s C0 is the instantaneous value of salt spray concentration, and A is the safe threshold for salt spray concentration. h A0 represents the humidity stress fluctuation amplitude, and T represents the permissible humidity stress amplitude. q T0 is the cumulative temperature difference cycle, T0 is the cumulative temperature difference cycle threshold, and min is the minimum value function.

[0085] It is important to know that the determination of the salt spray concentration safety threshold C0 is strictly based on the equipment anti-corrosion coating technical manual and ISO 12944 standard: for equipment with epoxy resin coating, C0 is set to 1.2 mg / m³; for polyurethane coating equipment, C0 is set to 0.5 mg / m³; if the equipment body is made of uncoated stainless steel, such as 316L stainless steel, then C0 is directly set to 3.0 mg / m³ based on the material's corrosion resistance level.

[0086] The determination of the permissible humidity stress amplitude A0 requires a comprehensive analysis of laboratory material testing and on-site environmental data. This can be achieved through accelerated aging tests, such as simulating a port environment according to IEC 60068-2-30 standards. Humidity cycling tests are then conducted on the metal structures of equipment, such as motor housings and control cabinet coatings or substrates, under constant temperature conditions. For example, the relative humidity is cycled between 30% and 85% and back to 30%, and the critical fluctuation amplitude A at which coating cracking or metal corrosion occurs is recorded. lab Further analysis of historical data from the site was conducted, and high-precision temperature and humidity sensors, such as the SHT85, were deployed to collect time-series data on the surface humidity of the equipment at a frequency of ≥1Hz. The moving standard deviation of humidity fluctuations over a continuous 24-hour period was calculated, and the 95th percentile value over 90 days was taken as the site baseline A. fieldFinally, calibration is performed using the formula:

[0087] ,

[0088] Where, k s For safety, the default value is 0.8; for example, the A-grade epoxy resin coating of a dust collector motor in a port... lab =18%RH, Site A field =15%RH, then A0=15×0.8=12%RH.

[0089] The temperature difference cycle accumulation threshold T0 is based on international standards, such as IEC 60068-2-14 thermal fatigue testing or ASTM E1049 cyclic stress analysis to simulate the temperature cycling environment of port equipment. For example, a daily variation range of -10℃ to 50℃ is set, and repeated temperature difference loads are applied to the critical materials of the equipment, such as the motor housing coating or metal substrate, until cracks or failures occur. The critical cumulative amount T that leads to material damage is recorded. lab Then extract T from 90 consecutive days q The 95th percentile value is used as the on-site safety baseline T. field Finally, the result is obtained through calculation using the formula:

[0090] ,

[0091] Where, k t For safety, the default value is 0.8.

[0092] Furthermore, in the above technical solution, the comprehensive equipment degradation model is specifically as follows:

[0093] ,

[0094] The calculation rules for dynamic weights ω1, ω2, and ω3 are as follows:

[0095] ,

[0096] ,

[0097] ,

[0098] Where I is the comprehensive equipment deterioration index, S is the structural health coefficient, D is the energy consumption deviation coefficient, R is the environmental corrosion rate coefficient, Δt is the continuous monitoring time interval, τ is the time decay constant, e is the natural constant, D(t) is the energy consumption deviation coefficient value at the current time t, D(t-Δt) is the energy consumption deviation coefficient value at the previous monitoring time t-Δt, R(t) is the environmental corrosion rate coefficient value at the current time t, and R(t-Δt) is the environmental corrosion rate coefficient value at the previous monitoring time t-Δt.

[0099] S4. The early warning decision terminal generates a dynamic early warning threshold based on the comprehensive equipment deterioration index. When the comprehensive equipment deterioration index exceeds the dynamic early warning threshold, a hierarchical early warning instruction set is triggered.

[0100] Furthermore, in the above technical solution, the generation mechanism of the dynamic early warning threshold is specifically as follows:

[0101] ,

[0102] Where T is the dynamic warning threshold, δ is the sensitivity coefficient, and μ I σ is the moving average of the overall equipment deterioration index. I γ is the moving standard deviation of the equipment degradation composite index, and γ is the baseline adjustment factor.

[0103] It should be noted that γ is the baseline adjustment factor, specifically... ;

[0104] The value of the sensitivity coefficient δ is determined according to the equipment type. For example, if it is a high-risk critical equipment, such as an oily wastewater treatment host, then δ=1.8 is set; if it is a medium-risk general equipment, such as a dust removal fan, then δ=1.5 is set; if it is a corrosion-sensitive equipment, such as a seawater cooling pump, then δ=1.3 is set; if it is a historically frequently faulty equipment, such as an old dust collector motor, then δ=1.2 is set.

[0105] Furthermore, in the above technical solution, the tiered early warning instruction set includes:

[0106] Emergency shutdown judgment command is recorded as Level 1 warning, critical maintenance command is recorded as Level 2 warning, and performance degradation warning command is recorded as Level 3 warning;

[0107] The classification rules for the Level 1 warning are as follows: ;

[0108] The classification rules for the Level II early warning are as follows: ;

[0109] The classification rules for the three-level early warning are as follows: ;

[0110] Where, σ I Let I be the moving standard deviation of the comprehensive equipment deterioration index, T be the dynamic early warning threshold, and I be the comprehensive equipment deterioration index.

[0111] S5. The dynamic monitoring terminal parses the hierarchical early warning instruction set and dynamically adjusts the sampling frequency and monitoring dimension parameters of the multi-source sensor network according to the early warning level.

[0112] Furthermore, in the above technical solution, the dynamic adjustment in step S5 includes:

[0113] When the warning level is 1, the sampling frequency of the vibration spectrum data is increased to 3 times the baseline value, and the monitoring dimension of the temperature sensor is added.

[0114] When the warning level is Level II, the sampling frequency of energy efficiency stream data will be increased to twice the baseline value, and the real-time calibration mode for corrosion data will be activated.

[0115] When the warning level is 3, the default sampling frequency is maintained, and the monitoring weight of the axial amplitude deviation of the spectrum data is increased.

[0116] S6, the operation and maintenance execution terminal generates equipment maintenance work orders based on the hierarchical early warning instruction set, and links port equipment to perform shutdown isolation, spare parts allocation and performance optimization operations.

[0117] Furthermore, in the above technical solution, the operation in step S6 includes:

[0118] When a Level 1 warning is issued, an emergency shutdown work order is generated, triggering the equipment's power isolation protocol.

[0119] When the warning level is level 2, a preventive maintenance work order is generated, and spare motors and desalination consumables are automatically allocated.

[0120] When a Level 3 warning is issued, a performance calibration work order is generated to adjust the reactive power compensation parameters of the equipment.

[0121] It should be noted that the implementation logic of the automatic allocation is that the operation and maintenance execution terminal queries the spare parts inventory status in real time through the warehouse database interface of the port equipment management system; if the spare motor inventory is sufficient, an allocation instruction is automatically generated to the port logistics scheduling system; if the inventory is lower than the safety threshold, the purchase order generation module is triggered simultaneously, and the emergency procurement process is initiated in conjunction with the supplier management system.

[0122] The adjustment of reactive power compensation parameters is achieved through the embedded control system of the port's intelligent power distribution cabinet, modifying the switching threshold parameters of the capacitor compensation cabinet. Specifically, this includes: increasing the target power factor value to above 0.95 and decreasing the reactive power ratio threshold η. ref Up to 5%.

[0123] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An information-based intelligent diagnosis and fault early warning method for the operating status of port environmental protection equipment, characterized in that, This includes environmental protection equipment data acquisition terminals, edge computing terminals, dynamic monitoring terminals, central analysis terminals, early warning and decision-making terminals, and operation and maintenance execution terminals, specifically including the following steps: S1. The environmental protection equipment data acquisition terminal collects multi-dimensional operational data from the port environmental protection equipment itself and its operating environment through a multi-source sensor network, forming a raw data set of equipment status; the raw data set of equipment status includes three-dimensional vibration spectrum data, energy efficiency flow data, and meteorological environmental corrosion data. The three-dimensional vibration spectrum data includes the maximum axial amplitude deviation, radial vibration frequency offset, and torque fluctuation standard deviation; the energy efficiency stream data includes the unit energy consumption output power, reactive power ratio, and motor efficiency decay slope; the meteorological environment corrosion data includes the instantaneous salt spray concentration, humidity stress fluctuation amplitude, and temperature difference cycle accumulation. S2. The edge computing terminal uses the Fourier transform algorithm to denoise the three-dimensional vibration spectrum data, the Kalman filter algorithm to dynamically calibrate the energy efficiency stream data, and the time series alignment algorithm to synchronize the meteorological and environmental corrosion data, generating a preprocessed time series data stream. S3. The central analysis terminal analyzes and preprocesses the time-series data stream to obtain the structural health coefficient, energy consumption deviation coefficient, and environmental corrosion rate coefficient. It then inputs the structural health coefficient, energy consumption deviation coefficient, and environmental corrosion rate coefficient into the equipment deterioration comprehensive model and outputs the equipment deterioration comprehensive index. The structural health coefficient is specifically: ， Where α is the amplitude tolerance factor, β is the frequency sensitivity factor, λ is the torque fluctuation gain factor, and δ max σ is the maximum axial amplitude deviation. v Let be the standard deviation of torque fluctuation, Δf be the radial vibration frequency offset, ln be the logarithm to the base e, and e be the natural constant. The energy consumption deviation coefficient is specifically: ， Among them, P e P0 is the power output per unit energy consumption, P0 is the rated operating condition reference power, and η is the power output per unit energy consumption. rea η represents the proportion of reactive power. ref k is the reactive power threshold. e denoted as the slope of motor efficiency decay, tan is the tangent function, and π is pi. The environmental corrosion rate coefficient is specifically: ， Among them, C s C0 is the instantaneous value of salt spray concentration, and A is the safe threshold for salt spray concentration. h A0 represents the humidity stress fluctuation amplitude, and T represents the permissible humidity stress amplitude. q is the cumulative amount of temperature difference cycle, T0 is the cumulative threshold of temperature difference cycle, and min is the minimum value function; S4. The early warning decision terminal generates a dynamic early warning threshold based on the comprehensive equipment deterioration index. When the comprehensive equipment deterioration index exceeds the dynamic early warning threshold, a hierarchical early warning instruction set is triggered. S5. The dynamic monitoring terminal parses the hierarchical early warning instruction set and dynamically adjusts the sampling frequency and monitoring dimension parameters of the multi-source sensor network according to the early warning level. S6, the operation and maintenance execution terminal generates equipment maintenance work orders based on the hierarchical early warning instruction set, and links with port equipment to perform shutdown isolation, spare parts allocation and performance optimization operations.

2. The information-based intelligent diagnosis and fault early warning method for the operating status of port environmental protection equipment according to claim 1, characterized in that, The comprehensive equipment degradation model is as follows: ， The calculation rules for dynamic weights ω1, ω2, and ω3 are as follows: ， ， ， Where I is the comprehensive equipment deterioration index, S is the structural health coefficient, D is the energy consumption deviation coefficient, R is the environmental corrosion rate coefficient, Δt is the continuous monitoring time interval, τ is the time decay constant, e is the natural constant, D(t) is the energy consumption deviation coefficient value at the current time t, D(t-Δt) is the energy consumption deviation coefficient value at the previous monitoring time t-Δt, R(t) is the environmental corrosion rate coefficient value at the current time t, and R(t-Δt) is the environmental corrosion rate coefficient value at the previous monitoring time t-Δt.

3. The information-based intelligent diagnosis and fault early warning method for the operating status of port environmental protection equipment according to claim 1, characterized in that, The mechanism for generating the dynamic early warning threshold is as follows: ， Where T is the dynamic warning threshold, δ is the sensitivity coefficient, and μ I σ is the moving average of the overall equipment deterioration index. I γ is the moving standard deviation of the equipment degradation composite index, and γ is the baseline adjustment factor.

4. The information-based intelligent diagnosis and fault early warning method for the operating status of port environmental protection equipment according to claim 1, characterized in that, The tiered early warning instruction set includes: Emergency shutdown judgment command is recorded as Level 1 warning, critical maintenance command is recorded as Level 2 warning, and performance degradation warning command is recorded as Level 3 warning; The classification rules for the Level 1 warning are as follows: ; The classification rules for the Level II early warning are as follows: ; The classification rules for the three-level early warning are as follows: ; Where, σ I Let I be the moving standard deviation of the comprehensive equipment deterioration index, T be the dynamic early warning threshold, and I be the comprehensive equipment deterioration index.

5. The information-based intelligent diagnosis and fault early warning method for the operating status of port environmental protection equipment according to claim 1, characterized in that, The dynamic adjustment in step S5 includes: When the warning level is 1, the sampling frequency of the vibration spectrum data is increased to 3 times the baseline value, and the monitoring dimension of the temperature sensor is added. When the warning level is Level II, the sampling frequency of energy efficiency stream data will be increased to twice the baseline value, and the real-time calibration mode for corrosion data will be activated. When the warning level is 3, the default sampling frequency is maintained, and the monitoring weight of the axial amplitude deviation of the spectrum data is increased.

6. The information-based intelligent diagnosis and fault early warning method for the operating status of port environmental protection equipment according to claim 1, characterized in that, The operations in step S6 include: When a Level 1 warning is issued, an emergency shutdown work order is generated, triggering the equipment's power isolation protocol. When the warning level is Level 2, a preventive maintenance work order is generated, and spare motors and desalination consumables are automatically allocated. When a Level 3 warning is issued, a performance calibration work order is generated to adjust the reactive power compensation parameters of the equipment.

Images