A method for collecting and recording fault data set of a multi-energy flow integrated energy production platform

Abstract

The application discloses a kind of multi-energy flow comprehensive energy production platform fault data set acquisition and record specification method.This method takes the typical fault type of electrical fluid coupling production process as the starting point, determines the main fault equipment and fault type for electricity area, gas area, liquid area respectively, and determines the measuring point that needs to be recorded accordingly;According to the highest effective frequency of target equipment and measuring point, the fastest and slowest dominant time constant of process, sensor bandwidth, anti-aliasing filter bandwidth and fault diagnosis algorithm window requirement, the data sampling rate and data record length of each measuring point are calculated;Finally, the measuring point operating data is saved as the CSV data file named with millisecond timestamp, and the information corresponding to each data file is recorded through independent element information record file.The application ensures that the recorded operating data can reflect the whole process characteristics of different industrial process before fault, fault occurrence, fault stabilization and fault recovery, and improves the completeness, accuracy and reusability of fault diagnosis data.

Description

Technical Field

[0001] This invention belongs to the technical field of multi-energy flow integrated energy production systems, industrial fault diagnosis, fault dataset construction and operation data acquisition and recording, and in particular relates to a standardized method for fault dataset acquisition and recording of multi-energy flow integrated energy production platforms for renewable energy power generation, water electrolysis hydrogen production, and hydrogen and carbon dioxide synthesis methanol coupled production processes. Background Technology

[0002] Against the backdrop of increasing demand for large-scale renewable energy consumption and green fuel production, electrohydraulic coupling processes, consisting of renewable energy power generation, water electrolysis for hydrogen production, and the synthesis of methanol / ammonia from hydrogen and carbon dioxide, are gradually becoming important energy conversion and utilization methods. These processes simultaneously include electrical power supply, power conversion, water electrolysis for hydrogen production, hydrogen compression and transportation, carbon dioxide supply, methanol / ammonia synthesis, and separation and purification, characterized by multiple equipment, multiple operating conditions, multiple time scales, and strong coupling.

[0003] Fault diagnosis is a crucial technical means to ensure the safe and stable operation of electro-hydraulic coupling systems. Since short circuits, open circuits, transformer winding faults, and hydrogen production power supply failures in the upstream electrical system affect the input power of the water electrolysis hydrogen production unit, and catalyst deactivation, membrane perforation, heat exchanger failure, and circulating pump failure in the electrolyzer affect hydrogen production and purity, while catalyst deactivation, heat exchanger failure, and distillation unit failure on the methanol / ammonia synthesis side affect methanol production and product quality, it is necessary to construct a fault operation dataset covering the electrical, gas, and liquid zones.

[0004] Existing fault operation datasets are typically constructed by first temporarily determining measurement points and then recording fault data. This approach can easily lead to a disconnect between measurement point settings and fault mechanisms, failing to prove that the key information required for a particular fault type has been collected. Furthermore, the sampling rate and recording duration for different partitions are often set based on experience, lacking a correspondence with the target fault frequency band, equipment time constant, sensor bandwidth, and algorithm window.

[0005] Furthermore, the fault processes in the electric, gas, and liquid zones differ significantly in dynamic timescales, signal frequency bands, types of observable variables, and fault propagation modes. Electric zone faults typically manifest as rapid abrupt changes in voltage and current waveforms, transient oscillations, harmonic distortions, and protection actions, with key fault information concentrated on millisecond to second timescales. Gas zone faults typically manifest as dynamic shifts in variables such as pressure, flow rate, temperature, composition, and purity, with fault responses influenced by gas compression, pipeline transportation, gas-liquid separation, and equipment control processes, and concentrated on second to minute timescales. Liquid zone faults typically manifest as slow changes in variables such as liquid flow rate, temperature, liquid level, product purity, and separation efficiency, with responses influenced by reactor thermal inertia, liquid phase residence time, heat exchange processes, and separation and purification processes, potentially lasting from several minutes to several hours. Therefore, if the same set of sampling rate, recording duration and measurement point configuration rules are used for the electrical, gas and liquid regions, it is easy to cause the loss of rapid transient information in the electrical region, insufficient recording of process response in the gas region or truncation of slow dynamic process in the liquid region, which in turn affects the integrity of fault samples and the training effect of diagnostic models.

[0006] The aforementioned issues directly impact fault diagnosis model training, fault location algorithm verification, and cross-regional data fusion analysis. Particularly for electro-hydraulic systems, transient faults in the electrical region typically require high sampling rates for recording, while thermal, fluid, and reaction processes in the gas and liquid regions are usually recorded at lower sampling rates over longer periods. Without a unified data recording standard, it becomes difficult to simultaneously meet the needs of transient electrical fault analysis and slow-variable process fault analysis.

[0007] Therefore, the present invention urgently needs to address the problem of how to establish a standardized method for collecting and recording fault datasets for multi-energy flow integrated energy production platforms, which can configure measurement points, sampling rates, and recording durations according to different fault characteristics, equipment dynamics, sensor response capabilities, and diagnostic algorithm requirements. This method should clearly define the data collection partitions, measurement point configurations, fault type numbers, fault parameters, recording durations, sampling rates, operating status labels, and data storage formats before forming the fault dataset. Summary of the Invention

[0008] The purpose of this invention is to propose a standardized method for collecting and recording fault datasets of multi-energy flow integrated energy production platforms, in order to solve the problem that there is a lack of unified standards for fault types, measurement point configurations, sampling rates, recording durations, and file traceability relationships in the existing construction process of fault datasets for integrated energy production platforms.

[0009] To achieve the above objectives, the technical solution adopted by this invention is as follows: This invention provides a standardized method for collecting and recording fault datasets of a multi-energy flow integrated energy production platform, comprising:

[0010] Identify typical fault types in each production process of the electro-hydraulic system in the integrated energy production platform. The production processes include at least the power supply and distribution and hydrogen production power process in the electric zone, the water electrolysis and hydrogen production and hydrogen transportation process in the gas zone, and the carbon dioxide hydrogenation and methanol separation and purification process in the liquid zone.

[0011] Based on the typical fault types, determine the faulty equipment, fault location, fault mechanism, and observable variables of the fault, and determine the measurement points that need to be recorded based on the observable variables of the fault.

[0012] For each measuring point and the dynamic characteristics of the corresponding faulty device, the data sampling rate and data recording duration are determined based on the highest effective frequency of the dynamic characteristics of the target faulty device, the process-dominant time constant, the sensor bandwidth, and the fault diagnosis algorithm window.

[0013] The measurement point data, operation status labels and file metadata are recorded according to a unified data structure. The measurement point data is saved as a data file named with a millisecond-level timestamp, and the fault details and sampling rate information corresponding to each data file are recorded in an independent metadata record file.

[0014] As a further improvement, the typical fault types are determined separately for electrical, gas, and liquid zones, and each typical fault type is assigned a unique fault number, with the normal operating state number being 0.

[0015] As a further improvement, determining the measurement points based on observable fault variables includes: identifying the equipment where the fault occurred; identifying the equipment state variables that changed due to the fault, including voltage, current, temperature, pressure, flow rate, composition, and purity; writing direct or indirect measurement points that can characterize the changes in the state variables into a measurement point configuration table; and recording the measurement point number, measurement point name, measurement point unit, equipment to which the measurement point belongs, zone to which the measurement point belongs, and sensor information for each measurement point.

[0016] As a further improvement, the data sampling rate is determined according to the following formula:

[0017]

[0018] in, Sampling rate, To characterize the maximum value of the main characteristic frequency of the fault process, For the comprehensive sampling margin coefficient, The fastest dominant time constant for the process corresponding to the measurement point. The minimum number of sampling points within the fastest dominant time constant; according to the Nyquist sampling theorem... ≥2, The range is 5 to 10.

[0019] As a further improvement, considering the combined effects of the anti-aliasing filter transition band, sensor bandwidth, and engineering margin, the engineering requirements are as follows: ≥5.

[0020] As a further improvement, the effective information bandwidth of the measuring point is determined according to the following formula:

[0021]

[0022] in, The effective information bandwidth of the measurement point, The effective frequency band for the fault process For the sensor's response bandwidth, This is the cutoff frequency for anti-aliasing filters or signal conditioning systems.

[0023] As a further improvement, the data recording duration is determined according to the following formula:

[0024]

[0025] in, Total recording duration The baseline duration before the fault. For material transfer, measurement or control delays, The slowest dominant time constant for the zone corresponding to the measurement point. For dynamic response coverage multiple, This is the margin required for fault recovery, protection actions, or algorithm windows.

[0026] As a further improvement, when the time-domain record of the electric region is performed for frequency-domain analysis, the frequency resolution satisfies:

[0027]

[0028] in, For frequency resolution, The length of the time-domain record.

[0029] As a further improvement, the data file is saved in CSV format, specifically YYYYMMDD_HHMMSS_fff.csv; the first line of the CSV file is the name of the measurement point channel, the first column is the serial number, the middle column is the data collected at each measurement point, and the last column is the running status label.

[0030] As a further improvement, each data file corresponds to one fault condition; the operating status label before the fault occurs and after the fault is recovered is 0, and the operating status label after the fault occurs but before recovery is the corresponding fault number.

[0031] As a further improvement, the independent metadata record file records the file name, file number, partition, faulty device, fault type, fault number, fault location, fault severity, fault duration, fault start and end time, operating conditions, measurement point list, sampling rate, recording duration, filter bandwidth, sensor response time, data unit, and remarks for each data file.

[0032] The core of this invention lies in standardizing the preparatory work before collecting fault datasets into four consecutive steps:

[0033] 1. Identify typical fault types in each production process of electro-hydraulic systems;

[0034] 2. Identify the measurement points that require data recording based on the determined typical fault types;

[0035] 3. Calculate the sampling rate and recording duration based on the dynamic characteristics of each measuring point and the corresponding faulty equipment;

[0036] 4. Specify the saving format for measurement point data, data files, and independent metadata record files.

[0037] By using formulas for data sampling rate, effective information bandwidth of measurement points, data recording duration, and frequency resolution, data sampling rate and data recording duration are no longer isolated empirical values, but are jointly determined by the dynamic characteristics of faulty equipment, the dynamics of measurement points, and the requirements of diagnostic algorithms.

[0038] Compared with the prior art, the beneficial effects of the present invention are:

[0039] The method provided by this invention can ensure the logical correspondence between fault types and measurement points, make the sampling rate and recording duration have an interpretable calculation basis, and incorporate short-term high-frequency electrical transient data and long-term low-frequency process data into the same dataset specification; at the same time, it can achieve traceable management of each sample through millisecond-level timestamp data files and independent metadata record files. Attached Figure Description

[0040] Figure 1 This is a flowchart illustrating the method for collecting and recording fault datasets of a multi-energy flow integrated energy production platform in this embodiment of the invention.

[0041] Figure 2 This is a process flow diagram of the offshore electricity-hydrogen-alcohol integrated energy production platform in this embodiment of the invention;

[0042] Figure 3 This is a schematic diagram of the configuration of zoned measurement points and sampling parameters of the integrated energy production platform in this embodiment of the invention;

[0043] Figure 4 This is a schematic diagram of the electrical zone measurement point configuration and fault location in Embodiment 1 of the present invention;

[0044] Figure 5 This is a schematic diagram of the gas zone measuring point configuration and fault location in Embodiment 2 of the present invention;

[0045] Figure 6 This is a schematic diagram of the liquid area measuring point configuration and fault location in Embodiment 3 of the present invention;

[0046] Figure 7 This is a schematic diagram of a typical fault type data CSV file format in an embodiment of the present invention. Detailed Implementation

[0047] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are merely 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 should fall within the scope of protection of the present invention.

[0048] The specific implementation process of this invention is as follows: Figure 1 As shown:

[0049] S1. Determine typical fault types:

[0050] Based on the electro-hydraulic production process, the fault objects are divided into electrical, gas, and liquid zones, and a list of typical fault types is established for each zone, with a unique fault number assigned to each type. Typical faults in the electrical zone include line short circuits, power outages, transformer winding displacement, transformer inter-turn short circuits, open circuits in hydrogen production power supply switching tubes, and DC bus capacitor breakdown. Typical faults in the gas zone include electrolyzer catalyst deactivation, electrolyzer membrane perforation, PEM heat exchanger failure, PEM circulating pump failure, and hydrogen compressor leakage. Typical faults in the liquid zone include fixed-bed catalyst deactivation, methanol / ammonia auxiliary heat exchanger failure, and methanol / ammonia distillation unit failure.

[0051] S2. Determine the measuring points based on the observable variables of the fault:

[0052] For each type of fault, analyze the faulty equipment, fault location, fault mechanism, and observable variables. The observable variables are used to determine the measurement points in reverse, forming a measurement point configuration table. The measurement point configuration table should include at least the measurement point number, measurement point name, unit, equipment to which it belongs, zone to which it belongs, sampling rate, sensor response time, and filter bandwidth.

[0053] S3. Determine the data sampling rate and data recording duration:

[0054] For electrical waveform measurement points, the sampling rate is calculated based on the highest target characteristic frequency and the comprehensive sampling margin coefficient; for process variables such as temperature, pressure, flow rate, composition and purity, the sampling interval is checked based on the fastest dominant time constant; for recording duration, it is calculated based on the pre-fault baseline, transmission delay, slowest dominant time constant and algorithm window.

[0055] S4. Record data files and independent metadata record files:

[0056] Each fault condition corresponds to a measurement point data file, which is named with a millisecond-level timestamp. A separate metadata record file is set up to record the detailed fault information, sampling parameters and measurement point list corresponding to the data file.

[0057] Taking the process flow of an offshore electricity-hydrogen-methanol integrated energy production platform as an example, the platform converts offshore wind and solar resources into liquid methanol fuel through three production stages: offshore wind and solar power generation (electricity zone), PEM electrolysis for hydrogen production (gas zone), and hydrogen plus carbon dioxide to methanol production (liquid zone). The specific production process of this platform is as follows: Figure 2 As shown, the specific implementation method of the present invention will be described below using this platform as an example.

[0058] As attached Figure 3 As shown, considering the differences in fault response time scale, signal frequency band, observable variable type, and sensor response capability among the electric, gas, and liquid zones in the multi-energy flow integrated energy production platform, corresponding data measurement points, data sampling rates, and data recording durations are configured for each production zone. The fault operation data collected from each zone are then summarized according to a unified data file format, status label rules, and metadata recording rules to form a production platform fault dataset with a unified structure, clear labels, and traceability.

[0059] Example 1: Electrical Fault Data Acquisition and Recording:

[0060] In this embodiment, the electrical zone is used to record fault operation data related to power supply and distribution and hydrogen production power supply.

[0061] S1. Determine the typical fault types of the electrical area:

[0062] Table 1 shows the typical fault types and numbers of the power area in this embodiment: the normal state is numbered 0, and the faults such as line short circuit, power interruption, transformer winding displacement, transformer inter-turn short circuit, hydrogen production power supply switch tube open circuit and hydrogen production power supply capacitor breakdown are respectively numbered accordingly. Table 1

[0063]

[0064] S2. Determine the measuring points based on the observable variables of the fault:

[0065] Based on typical fault types and faulty equipment, the electrical area measurement points are configured as shown in Table 2, with a total of 38 measurement points set up. Table 2

[0066]

[0067] The locations of the power area measurement points and faulty equipment are shown in the attached figure. Figure 4 As shown;

[0068] S3. Determine the data sampling rate and data recording duration:

[0069] The electrical area mainly involves voltage, current, short-circuit transients, harmonics, transformer faults, and power switch malfunctions. It is necessary to pay attention to the higher harmonics of current and voltage during fault processes. The value range of the sampling rate calculation parameters for the electrical area is shown in Table 3. Table 3

[0070]

[0071] Frequency constraints:

[0072]

[0073] Substitute:

[0074]

[0075] so:

[0076]

[0077] Time constant constraint:

[0078]

[0079] Substitute:

[0080]

[0081] Approximately:

[0082]

[0083] Therefore, the comprehensive recommendations for power areas are shown in Table 4: Table 4

[0084]

[0085] Therefore, 10000 Hz can be used as a preferred sampling frequency to cover the 50 Hz fundamental frequency, the 50th harmonic, and the main low-frequency fault transients.

[0086] For a 50 Hz system, one power frequency cycle is 20 ms. When the recording duration is set to 1 s, 50 complete power frequency cycles can be obtained, with a frequency resolution of 1 Hz. When the fault occurs at 0.10 to 0.11 s, about 5 complete cycles before the fault can be obtained. After the fault occurs, about 44 to 45 cycles can still be obtained. This can cover the fault formation, protection action, fault clearing, and short-term recovery process. Therefore, the 1 s fault process recording duration meets the requirements for electrical area fault analysis.

[0087] S4. Record data files and independent metadata record files:

[0088] In this embodiment, the short-circuit faults of the 35kV submarine cable line include single-phase grounding of phases A, B, and C; two-phase short circuits between phases AB, BC, and AC; two-phase grounding short circuits between phases AB, BC, and AC; three-phase short circuits of phases ABC; and three-phase grounding short circuits of phases ABC. The grounding resistance is set to 0.1Ω to 100Ω, the fault duration is set to 0.02s to 0.9s, where 0.9s indicates that the fault has not been recovered within the simulation recording time, and the operating condition is set to 0.1pu to 1.2pu.

[0089] The faults of 35kV submarine cable power supply interruption, 10kV AC bus power supply interruption and 380V auxiliary power supply interruption include single-phase circuit breakers of phases A, B and C, two-phase circuit breakers of phases AB, BC and AC, and three-phase circuit breakers of phases ABC; the fault duration is set to 0.02s to 0.9s, and the operating condition is set to 0.1pu to 1.2pu.

[0090] The 35kV / 10kV step-down transformer winding displacement fault includes the displacement of phase A, phase B, and phase C windings. The fault is non-recoverable after it occurs, and the fault severity is set to 5% to 50%. The operating condition is set to 0.1pu to 1.2pu. The 35kV / 10kV step-down transformer inter-turn short circuit fault includes the inter-turn short circuit of phase A, phase B, and phase C. The fault is non-recoverable after it occurs, and the fault severity is set to 1% to 20%. The operating condition is set to 0.1pu to 1.2pu.

[0091] For open-circuit faults in the hydrogen production power supply switching transistors, different open-circuit conditions are selected based on the hydrogen production power supply topology. Once a fault occurs, it is not recoverable, and the operating condition is set to 0.1 pu to 1.2 pu. For capacitor breakdown faults in the hydrogen production power supply, the fault severity is set to 5% to 50% by reducing the DC bus capacitor value. Once a fault occurs, it is not recoverable, and the operating condition is set to 0.1 pu to 1.2 pu.

[0092] The electrical data was recorded for a total of 1 second. The fault start time was randomly set between 0.1s and 0.11s within the recording interval, i.e., between the 1000th and 1100th sampling points. The data was stored in CSV format, with the file name format YYYYMMDD_HHMMSS_fff. The first row was the measurement point channel name, the first column was the serial number, and the last column was the operating status. The data file format is shown in the attached figure. Figure 7 As shown.

[0093] The fault details for each file were recorded, resulting in the following fault information correspondence table 5: Table 5

[0094]

[0095] Example 2: Gas Area Fault Data Acquisition and Recording

[0096] In this embodiment, the gas zone is used to record fault operation data related to the PEM electrolyzer, hydrogen production auxiliary equipment, and hydrogen compression and transportation.

[0097] S1. Determine the typical fault types in the gas area:

[0098] Table 6 shows the typical fault types and numbers in the gas area in this embodiment.

[0099] Table 6

[0100]

[0101] S2. Determine the measuring points based on the observable variables of the fault:

[0102] Based on typical fault types and faulty equipment, the gas area measurement points are configured as shown in Table 7, with a total of 19 measurement points set up.

[0103] Table 7

[0104]

[0105] The locations of the gas area data measurement points and faulty equipment are shown in the attached figure. Figure 5 As shown.

[0106] S3. Determine the data sampling rate and data recording duration:

[0107] The gas zone mainly contains variables such as pressure, flow rate, temperature, composition, purity, and leakage of hydrogen, oxygen, carbon dioxide, circulating gas, and mixed gas, as shown in Table 8.

[0108] Table 8

[0109]

[0110] Frequency constraints:

[0111]

[0112] so:

[0113]

[0114] Time constant constraint:

[0115]

[0116] so:

[0117]

[0118] Therefore, the comprehensive recommendations for air zones are shown in Table 9: Table 9

[0119]

[0120] Therefore, the sampling rate for the air zone is specified as 1 Hz, and the recording duration is 1000 s.

[0121] S4. Record data files and independent metadata record files:

[0122] The catalyst deactivation fault in the electrolyzer is controlled by reducing the anode catalyst exchange current density, with the fault severity set to 5% to 50%. Once the fault occurs, it is irreversible, and the operating condition is set to 0.05 pu to 1.5 pu. The membrane perforation fault in the electrolyzer is controlled by reducing the number of series-connected cells and lowering the Faraday efficiency. Once the fault occurs, it is irreversible, and the operating condition is set to 0.05 pu to 1.5 pu.

[0123] The PEM heat exchanger failure fault is set by reducing the cooling fan speed to 0, with the fault duration set from 10s to 900s, where 900s indicates no recovery after the fault occurs, and the operating condition is set from 0.1pu to 1.2pu. The PEM circulating pump failure fault is set by reducing the circulating pump speed to 0, with the fault duration set from 10s to 900s, where 900s indicates no recovery after the fault occurs, and the operating condition is set from 0.1pu to 1.2pu.

[0124] The hydrogen zone data was recorded for a total of 1000 seconds. The fault start time was randomly set between 100 and 110 seconds within the recording interval, i.e., between the 100th and 110th sampling points. The data was stored in CSV format, with the file name format YYYYMMDD_HHMMSS_fff. The first row is the measurement point channel name, the first column is the serial number, and the last column is the operating status. The data file format is attached. Figure 7 As shown.

[0125] The fault information for each file was recorded, resulting in the following fault information correspondence table as shown in Table 10: Table 10

[0126]

[0127] Example 3: Data Acquisition and Recording of Liquid Zone Faults

[0128] In this embodiment, the liquid zone is used to record fault operation data of hydrogen, carbon dioxide, circulating gas, mixed gas, fixed-bed reactor, and methanol / ammonia distillation process.

[0129] S1. Determine the typical fault types in the liquid area:

[0130] Table 11 shows the typical fault types and numbers in the liquid zone in this embodiment.

[0131] Table 11

[0132]

[0133] S2. Determine the measuring points based on the observable variables of the fault:

[0134] Based on typical fault types and faulty equipment, the liquid area measuring points are configured as shown in Table 12, with a total of 24 measuring points set up.

[0135] Table 12

[0136]

[0137] The data points for the liquid area and the location of the faulty equipment are shown in the attached figure. Figure 6 As shown.

[0138] S3. Determine the data sampling rate and data recording duration:

[0139] The liquid zone mainly consists of variables such as the methanol liquid phase, the streams before and after distillation, product purity, heat exchange, liquid level, and liquid flow rate. The liquid zone is usually slower than the gas zone, as shown in Table 13.

[0140] Table 13

[0141]

[0142] Frequency constraints:

[0143]

[0144] so:

[0145]

[0146] Time constant constraint:

[0147]

[0148] so:

[0149]

[0150] Therefore, the comprehensive recommendations for liquid storage areas are shown in Table 14: Table 14

[0151]

[0152] Therefore, the sampling rate for the liquid area is specified as 1 Hz, and the recording time is 4000 s.

[0153] S4. Record data files and independent metadata record files:

[0154] Fixed-bed catalyst deactivation faults characterize the performance degradation caused by decreased catalyst activity in the methanol synthesis reactor, with fault severity set to 5% to 50%, and no recovery after occurrence. Methanol auxiliary heat exchanger failure faults characterize temperature deviations caused by decreased or failed heat exchange capacity during methanol synthesis and separation, with operating conditions set to 0.1 pu to 1.2 pu. Methanol distillation unit failure faults characterize the impact of the distillation process on methanol product purity and flow rate.

[0155] Data for the liquid area was recorded for a total of 4000 seconds. The fault start time was randomly set between 100 and 110 seconds within the recording interval, i.e., between the 100th and 110th sampling points. The data was stored in CSV format, with the file name format YYYYMMDD_HHMMSS_fff. The first row is the measurement point channel name, the first column is the serial number, and the last column is the operating status. The data file format is shown in the attached figure. Figure 7 As shown.

[0156] The fault details of each document were recorded, resulting in the following table (Table 15) showing the corresponding fault information for the liquid area: Table 15

[0157]

[0158] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the invention.

[0159] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

[0160] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any other way. Any modifications or equivalent changes made based on the technical essence of the present invention shall still fall within the scope of protection claimed by the present invention.

Claims

1. A standardized method for collecting and recording fault datasets of a multi-energy flow integrated energy production platform, characterized in that, include: Identify typical fault types in each production process of the electrical and hydraulic systems in the integrated energy production platform. The production processes include at least the power supply and distribution and hydrogen production power process in the electrical zone, the water electrolysis and hydrogen production and hydrogen transportation process in the gas zone, and the carbon dioxide hydrogenation and methanol separation and refining process in the liquid zone. Based on the typical fault types, determine the faulty equipment, fault location, fault mechanism, and observable variables of the fault, and determine the measurement points that need to be recorded based on the observable variables of the fault. For each measuring point and the dynamic characteristics of the corresponding faulty device, the data sampling rate and data recording duration are determined based on the highest effective frequency of the dynamic characteristics of the target faulty device, the process-dominant time constant, the sensor bandwidth, and the fault diagnosis algorithm window. The measurement point data, operation status labels and file metadata are recorded according to a unified data structure. The measurement point data is saved as a data file named with a millisecond-level timestamp, and the fault details and sampling rate information corresponding to each data file are recorded in an independent metadata record file.

2. The method for collecting and recording fault datasets of a multi-energy flow integrated energy production platform according to claim 1, characterized in that, The typical fault types are determined separately for electrical, gas, and liquid zones, and each typical fault type is assigned a unique fault number, with the normal operating state number being 0.

3. The method for collecting and recording fault datasets of a multi-energy flow integrated energy production platform according to claim 1, characterized in that, Determining measurement points based on observable fault variables includes: identifying the equipment where the fault occurred; identifying the equipment state variables that changed due to the fault, including voltage, current, temperature, pressure, flow rate, composition, and purity; writing direct or indirect measurement points that can characterize the changes in the state variables into a measurement point configuration table; and recording the measurement point number, measurement point name, measurement point unit, equipment to which the measurement point belongs, zone to which the measurement point belongs, and sensor information for each measurement point.

4. The method for collecting and recording fault datasets of a multi-energy flow integrated energy production platform according to claim 1, characterized in that, The data sampling rate is determined according to the following formula: ； in, Sampling rate, To characterize the maximum value of the characteristic frequency of the fault process, For the comprehensive sampling margin coefficient, The fastest dominant time constant for the process corresponding to the measurement point. The minimum number of sampling points within the fastest dominant time constant; according to the Nyquist sampling theorem... ≥2, The range is 5 to 10.

5. The method for collecting and recording fault datasets of a multi-energy flow integrated energy production platform according to claim 1, characterized in that, The effective information bandwidth of the measuring point is determined according to the following formula: ； in, The effective information bandwidth of the measurement point, The effective frequency band for the fault process For the sensor's response bandwidth, This is the cutoff frequency for anti-aliasing filters or signal conditioning systems.

6. The method for collecting and recording fault datasets of a multi-energy flow integrated energy production platform according to claim 1, characterized in that, The data recording duration is determined according to the following formula: ； in, Total recording duration The baseline duration before the fault. For material transfer, measurement or control delays, The slowest dominant time constant for the zone corresponding to the measurement point. For dynamic response coverage multiple, This is the margin required for fault recovery, protection actions, or algorithm windows.

7. The method for collecting and recording fault datasets of a multi-energy flow integrated energy production platform according to claim 4, characterized in that, When performing frequency domain analysis on the time-domain record of the electric region, the frequency resolution satisfies: ； in, For frequency resolution, The length of the time-domain record.

8. The method for collecting and recording fault datasets of a multi-energy flow integrated energy production platform according to claim 1, characterized in that, The data file is saved in CSV format, specifically YYYYMMDD_HHMMSS_fff.csv. The first line of the CSV file contains the name of the measurement point channel, the first column contains the serial number, the middle column contains the data collected at each measurement point, and the last column contains the running status label.

9. The method for collecting and recording fault datasets of a multi-energy flow integrated energy production platform according to claim 8, characterized in that, Each data file corresponds to a fault condition; the operating status label before the fault occurs and after the fault is recovered is 0, and the operating status label after the fault occurs but before recovery is the corresponding fault number.

10. The method for collecting and recording fault datasets of a multi-energy flow integrated energy production platform according to claim 1, characterized in that, The independent metadata record file records the file name, file number, partition, faulty device, fault type, fault number, fault location, fault severity, fault duration, fault start and end time, operating conditions, list of measurement points, sampling rate, recording duration, filter bandwidth, sensor response time, data unit, and remarks for each data file.

Images