Converter valve operation state evaluation method, system and equipment based on static health and dynamic trend dual-path fusion, and medium

Abstract

The invention belongs to the technical field of modular multi-level converter valve evaluation, and discloses a converter valve operation state evaluation method, system, equipment and medium based on static health and dynamic trend dual-path fusion, so as to overcome the limitation that only static state evaluation is emphasized in the prior art. The method comprises the following steps: acquiring electrical parameters, thermal parameters and mechanical and environmental parameters reflecting the operation state of the converter valve in real time as multi-dimensional evaluation factors; based on the multi-dimensional evaluation factor, executing the static health degree evaluation path and the dynamic trend degree evaluation path in parallel to obtain a static health index and a dynamic trend index; and inputting the static health index and the dynamic trend index into a preset two-dimensional state space for analysis, and mapping and outputting a final operation state grade of the converter valve. According to the invention, early warning and predictive maintenance of performance degradation are realized, the predictive capability of the state evaluation system is improved, and safe and stable operation of the flexible DC power transmission system is guaranteed.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of modular multilevel converter evaluation, and particularly relates to a method, system, device and medium for evaluating the operating state of a converter based on the fusion of static health and dynamic trend. BACKGROUND

[0002] Modular multilevel converter (MMC) is the core equipment in flexible DC transmission system, and the reliability of its operating state has a decisive influence on the safety and stability of the entire transmission system. Therefore, developing advanced online state evaluation technology to realize early warning and predictive maintenance of performance degradation of the converter has become a key issue in this technical field.

[0003] At present, the state evaluation methods of the converter can be mainly divided into two categories. The first category of method is based on fixed threshold criterion, such as the scheme proposed in Chinese invention patent (publication number CN111596160A), which monitors key electrical quantities such as bridge arm current and sub-module capacitor voltage in real time, calculates derived parameters such as capacitor capacity and IGBT junction temperature, and then compares the measured values with the preset safety threshold to determine whether the equipment is in a fault state. This method can effectively identify sudden hard failures, but it cannot quantitatively characterize the progressive performance degradation of the equipment, nor does it have the ability of forward-looking warning. To provide more comprehensive state awareness, the second category of method adopts a multi-factor comprehensive scoring strategy. For example, Chinese invention patent (publication number CN117516917A) constructs multi-dimensional evaluation factors and calculates health scores combined with weights to realize quantitative evaluation of the overall health level of the converter, and usually has independent heavy fault judgment logic to ensure system safety. Compared with the threshold method, it has improved the degree of refinement in state description. However, the above-mentioned prior art solutions are essentially static state evaluation category. The evaluation results are heavily dependent on the instantaneous data at the current time, and the core purpose is to answer "what is the current state of the equipment", while for the key question "how is the future trend of the health state of the equipment", there is a lack of effective analysis means. In actual operation, when the key parameters (such as IGBT junction temperature) have not exceeded the limit but have shown an accelerating degradation trend, the purely static evaluation model is difficult to identify such potential high-risk working conditions, which may miss the best early intervention opportunity.

[0004] In summary, the existing technology is limited to the static description of the current health level of the equipment, and cannot effectively fuse dynamic trend information to realize forward-looking risk warning, which restricts the further improvement of the prediction ability of the state evaluation system. SUMMARY

[0005] Based on the above-mentioned shortcomings and deficiencies existing in the prior art, one of the purposes of the present application is to at least solve one or more of the above-mentioned problems existing in the prior art, in other words, one of the purposes of the present application is to provide a static health and dynamic trend dual-path fusion based converter valve operating state evaluation method, system, device and medium to meet one or more of the aforementioned needs, to realize early warning and predictive maintenance of performance degradation, improve the prediction ability of the state evaluation system, and ensure the safe and stable operation of the flexible DC power transmission system.

[0006] In order to achieve the above-mentioned purposes of the application, the following technical solutions are adopted in the present application: In a first aspect, the present application provides a static health and dynamic trend dual-path fusion based converter valve operating state evaluation method, comprising the steps of: S1, real-time acquisition of electrical parameters, thermal parameters, and mechanical and environmental parameters reflecting the operating state of the converter valve as multi-dimensional evaluation factors; S2, based on the multi-dimensional evaluation factors, performing a static health degree evaluation path and a dynamic trend degree evaluation path in parallel; The static health degree evaluation path is used to calculate a static health index representing the current comprehensive health level of the converter valve according to the numerical values of the multi-dimensional evaluation factors at the current time through a preset scoring model and a weighting algorithm; The dynamic trend degree evaluation path is used to calculate a dynamic trend index representing the change trend of the health state of the converter valve according to the historical data sequence of the multi-dimensional evaluation factors in the past period of time through a trend analysis algorithm; S3, inputting the static health index and the dynamic trend index into a preset two-dimensional state space for analysis, mapping and outputting the final operating state grade of the converter valve.

[0007] As a preferred scheme, a highest priority serious fault judgment step is performed before or in parallel with step S2: Real-time monitoring of whether there is a preset serious fault trigger condition; If any of the serious fault trigger conditions is met, the subsequent evaluation process is interrupted, and a serious fault state is directly output.

[0008] As a preferred scheme, the static health degree evaluation path is performed specifically as follows: The objective initial weight of each evaluation factor is calculated by entropy weight method; The standardized numerical value of each evaluation factor is processed by fuzzy membership function to obtain the health score of each evaluation factor corresponding to different health grades; Based on the objective initial weight, the health scores of each evaluation factor are weighted and summed to calculate the static health index.

[0009] As a preferred scheme, the fuzzy membership function is a triangular membership function for defining three fuzzy sets of high health, medium health and low health and their membership degrees.

[0010] As a preferred scheme, the trend analysis algorithm in the dynamic trend degree evaluation path comprises the following steps: performing smoothing and noise reduction preprocessing on the historical data sequence; fitting the preprocessed data using a Holt-Winters exponential smoothing model to calculate a level component, a trend component and a prediction value; based on the trend component and / or the prediction value, quantifying the severity of degradation and converting it into the dynamic trend index.

[0011] As a preferred scheme, the smoothing parameters a and β of the Holt-Winters exponential smoothing model have a value range of 0 < a, β < 1, and the prediction step h = 1.

[0012] As a preferred scheme, the two-dimensional state space is a two-dimensional state fusion matrix, with the static health index as the first dimension coordinate and the dynamic trend index as the second dimension coordinate, and different regions of the two-dimensional state fusion matrix are pre-defined with different operating state levels.

[0013] In a second aspect, the present application provides a converter valve operating state evaluation system based on static health and dynamic trend two-path fusion, for implementing the converter valve operating state evaluation method as described in the first aspect, comprising: a data acquisition module for acquiring electrical parameters, thermal parameters, and mechanical and environmental parameters reflecting the operating state of the converter valve in real time as multi-dimensional evaluation factors; a data processing and evaluation module including a static evaluation unit, a dynamic evaluation unit and a fusion diagnosis unit, for performing a static health degree evaluation path, a dynamic trend degree evaluation path and a state fusion diagnosis; the static evaluation unit is used to calculate a static health index; the dynamic evaluation unit is used to calculate a dynamic trend index; the fusion diagnosis unit is used to output an operating state level in a two-dimensional state space based on the static health index and the dynamic trend index; a severe fault judgment module for performing severe fault judgment with the highest priority; a data storage module for storing historical data sequences of the multi-dimensional evaluation factors.

[0014] In a third aspect, the present application provides an electronic device, the computer device comprising a memory, a processor and a computer program, the computer program being executed by the processor to implement the converter valve operating state evaluation method as described in the first aspect.

[0015] In a fourth aspect, the application provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the converter valve operating state evaluation method according to the first aspect.

[0016] Compared with the prior art, the application has the following beneficial effects: 1. A leap from static evaluation to predictive management is achieved: By introducing an independent dynamic trend degree evaluation path, the application overcomes the limitation of the prior art that can only reflect the current static health level of the equipment. This method not only answers "how is the current state of the equipment", but also effectively evaluates "how is the state of the equipment changing", thereby realizing early warning and predictive maintenance of the performance degradation trend, helping the operation and maintenance personnel to seize the best maintenance window.

[0017] 2. The comprehensiveness and accuracy of state evaluation are improved: By constructing an evaluation framework of "static health index S_H" and "dynamic trend index S_T" dual-path parallel fusion, the application can more finely distinguish different risk states. In particular, it can accurately identify the implicit high-risk state (hidden danger warning) of "good current state but rapid degradation trend" and the chronic low-risk state of "poor current state but stable", providing unprecedented scientific basis for differentiated and precise operation and maintenance decisions.

[0018] 3. An intelligent evaluation mechanism combining objectivity and self-adaptation is adopted: In the static health degree evaluation, the entropy weight method and the fuzzy membership function are innovatively combined. The entropy weight method objectively assigns weights based on the variability of the data itself, avoiding the subjectivity of manually setting weights; the fuzzy logic effectively handles the uncertainty problem of parameter boundaries, making the health score more consistent with engineering practice, and significantly improving the objectivity and reliability of the evaluation results.

[0019] 4. The real-time and safety of system response are ensured: A highest priority serious fault judgment step is set, which runs in parallel with the core evaluation process. Once the trigger condition is monitored, the complex calculation can be interrupted immediately and an alarm can be given directly, ensuring the rapid response capability to sudden serious faults, thereby ensuring the safe and stable operation of the converter valve and the entire power transmission system.

[0020] 5. Intuitive and efficient decision support is provided: The two-dimensional state fusion matrix is used to fuse and visualize the dual-path evaluation results, mapping complex multi-dimensional information into intuitive operating state levels. This greatly reduces the threshold of result interpretation, making it easy for operation and maintenance personnel to quickly and accurately grasp the overall health status of the equipment and develop appropriate strategies, thereby improving operation and maintenance efficiency.

[0021] Further or more detailed beneficial effects will be described in the specific embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed in the embodiments or prior art description. Obviously, the drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative effort.

[0023] Figure 1 is a flowchart of the converter valve operating state evaluation method according to the first embodiment of the present application.

[0024] Figure 2 is a structural diagram of the converter valve operating state evaluation system according to the second embodiment of the present application.

[0025] Figure 3 is a structural diagram of the electronic device according to the embodiment of the present application.

[0026] Reference Signs: 300, electronic device; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory. DETAILED DESCRIPTION

[0027] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application.

[0028] In the following description, a plurality of embodiments of the present application are provided, and different embodiments can be replaced or combined, so the present application can also be considered to include all possible combinations of the same and / or different embodiments described. Thus, if one embodiment includes features A, B, and C, and another embodiment includes features B and D, the present application should also be considered to include embodiments including one or more of all other possible combinations of A, B, C, and D, even if the embodiment is not explicitly described in the following content.

[0029] The following description provides examples and does not limit the scope, applicability, or examples set forth in the claims. Changes can be made in the function and arrangement of elements described without departing from the scope of the present application. Various examples can omit, substitute, or add various procedures or components as appropriate. For instance, the methods described can be performed in an order different than described, and various steps can be added, omitted, or combined. Also, features described with respect to some examples can be combined in other examples.

[0030] In order to facilitate better understanding of the embodiments of the present application, before the specific embodiments of the present application are explained in detail, the application scenarios thereof will be described.

[0031] The converter valve operating state evaluation method described in the embodiments of the present specification is applied to VSC-HVDC power transmission projects and other flexible DC power transmission systems using modular multilevel converter valves. In these scenarios, the application of the converter valve operating state evaluation method aims to accurately evaluate and early warn the operating state of the converter valve by comprehensively integrating the current static health level and future dynamic degradation trend information of the device, thereby improving the reliability and economy of system operation and effectively avoiding unexpected downtime.

[0032] The static health degree evaluation path, dynamic trend degree evaluation path, trend analysis algorithm, two-dimensional state space, fuzzy membership function, and Holt-Winters exponential smoothing model involved in the embodiments of the present specification are briefly explained as follows: Static health degree evaluation path: Focuses on the comprehensive health state evaluation of the converter valve at the current time. By real-time collection of electrical parameters (such as bridge arm current, sub-module capacitor voltage, etc.), thermal parameters (such as IGBT junction temperature, cooling water temperature, etc.), and mechanical and environmental parameters (such as vibration signal, IGBT module thermal resistance, etc.) of the converter valve, and using the preset static scoring model and weighting algorithm (such as entropy weight method combined with fuzzy membership function), the static health index S_H representing the current health level of the converter valve is calculated. This path provides the device's real-time health state information for the operation and maintenance personnel.

[0033] Dynamic trend degree evaluation path: Unlike static evaluation, the dynamic trend degree evaluation path focuses on the change trend of the converter valve health state in the future. It analyzes the historical data sequence of multi-dimensional evaluation factors and uses trend analysis algorithms (such as smoothing and denoising processing combined with Holt-Winters exponential smoothing model) to predict the future changes of the converter valve health state and calculate the dynamic trend index S_T. This path helps to discover potential degradation trends in advance and provides a basis for predictive maintenance.

[0034] Trend analysis algorithm is the core part of the dynamic trend degree evaluation path, which is used to process historical data sequences to reveal their change trends. It usually includes two steps of data smoothing and denoising preprocessing and trend fitting. Smoothing and denoising aims to eliminate random fluctuations in the data, while trend fitting captures the long-term change trend of the data through mathematical models (such as Holt-Winters model).

[0035] The two-dimensional state space is a state evaluation tool that combines static health index (S_H) and dynamic trend index (S_T). It constructs a two-dimensional matrix with S_H as the horizontal axis and S_T as the vertical axis. Different regions of the matrix represent different operating state levels (such as healthy, sub-healthy, hidden danger warning, and failure). By mapping the calculated S_H and S_T values to the matrix, the final operating state level of the converter valve can be determined intuitively.

[0036] The fuzzy membership function is an important tool for handling parameter uncertainty and nonlinear boundaries in the static health evaluation path. It defines fuzzy sets for different health levels (such as high health, medium health, and low health) and corresponding membership functions to convert continuous parameter values into discrete health scores. This method avoids the rigidity of binary judgments and enables fine quantization of multi-level health ratings.

[0037] The Holt-Winters exponential smoothing model is a time series prediction method widely used in the dynamic trend evaluation path. This model considers the level, trend, and seasonal components of the data to fit and predict historical data. In the evaluation of converter valve operating states, the Holt-Winters model is used to capture long-term trends and short-term fluctuations in historical data of multi-dimensional evaluation factors, accurately predicting future changes in the health status of the converter valve.

[0038] Embodiment One: As shown in Figure 1 , the embodiment provides a converter valve operating state evaluation method based on the fusion of static health and dynamic trend. It is specifically applied to VSC-HVDC transmission projects.

[0039] In this project, the converter valve adopts a modular multilevel structure, with each bridge arm composed of N sub-modules. The system is equipped with a valve control system, a station control system, and a sensor network to realize real-time monitoring. The evaluation system is built based on a digital processor, with software written in VHDL language and parallel processing capability. The data sampling frequency is set to 100 kHz, and the collected historical data is stored in a local database for subsequent analysis.

[0040] The converter valve operating state evaluation method includes the following key steps: Step S1: Multi-dimensional data acquisition: Real-time acquisition of multi-dimensional evaluation factors reflecting the operating state of the converter valve.

[0041] Specifically, the multi-dimensional evaluation factors include: 1. Electrical parameters include bridge arm current, sub-module capacitor voltage, IGBT drive state signal, IGBT switching frequency, bypass switch state, and sub-module operating power supply state; 2. Thermal parameters include IGBT junction temperature, cooling water inlet and outlet temperature obtained from valve control system (VBC) or station control system (SCADA), ambient temperature; 3. Mechanical and environmental parameters include vibration signal features of key components (e.g. collected by acceleration sensors, used to find mechanical problems such as bolt loosening, structural fatigue, etc.).

[0042] Before or in parallel with step S2, a step S is performed, which is a severe fault judgment: Real-time monitoring whether there is a preset severe fault trigger condition, such as IGBT drive signal loss, communication interruption, capacitor voltage exceeding 150% of the rated value; If any of the severe fault trigger conditions is met, all subsequent evaluation processes are immediately interrupted, and a "severe fault" state is directly output and an alarm is triggered.

[0043] Step S2 is a double-path parallel evaluation: If no severe fault is triggered, based on the multi-dimensional evaluation factors, the following two evaluation paths are executed in parallel: Path A: Static health degree evaluation This path is used to calculate the static health index S_H representing the current comprehensive health level of the converter valve, and its specific steps include: 1. Data standardization and weight calculation: standardize the original measurement values of each evaluation factor, and objectively calculate the initial weight of each factor based on the variability of historical data using entropy weight method.

[0044] 2. Fuzzy membership degree evaluation: define three fuzzy sets of high health, medium health and low health using triangular membership functions, calculate the membership degree of each factor standardized value to different health levels, and convert it to specific health score.

[0045] 3. Index synthesis: based on the initial weight, the health score of each evaluation factor is weighted and summed to finally obtain the static health index S_H.

[0046] Suppose the current time (t=0) data collected in step S1 is as follows (unit: standardization): 1. Electrical parameters: (1) Bridge arm current: 500A (normal range: 0A-800A); (2) Sub-module capacitor voltage: 1500V (normal: 1400V-1600V); (3) IGBT drive state signal: normal (0 / 1 binary); (4) IGBT switching frequency: 250Hz (normal: 200Hz-300Hz); (5) bypass switch status: closed (normal); (6) sub-module operating power supply status: normal (voltage stabilized at 24V); 2. Thermal parameters: (1) IGBT junction temperature: 80°C (normal: <100°C, obtained from VBC); (2) Cooling water inlet and outlet temperature: inlet water 30°C, outlet water 40°C (obtained from SCADA); (3) Ambient temperature: 25°C (obtained from SCADA); 3. Mechanical and environmental parameters: (1) Vibration signal characteristics: amplitude 0.5g (normal: <1g, collected by acceleration sensor); (2) Additional parameters: (3) Real-time thermal resistance of IGBT module Rth(j-c): online estimation is 0.05K / W (based on thermal model, normal: <0.1K / W).

[0047] Then the calculation process of the static health index S_H based on the above data at t=0 includes: First step: data standardization and entropy weight calculation Convert the original measurement value to a dimensionless standardized value [0, 1] (1 represents the best health state), and then use the entropy weight method to objectively determine the weight of each factor. Entropy weight method is based on information theory: factors with high variability (i.e. high contribution to health assessment) have higher weights because they have more "information".

[0048] 1. Data standardization: For positive indicators (the higher the better, such as stability of bridge arm current within normal range), the standardization formula is: , where, is the value of the i th factor of the j th sample.

[0049] For negative indicators (the lower the better, such as IGBT junction temperature, vibration, thermal resistance), the standardization formula is: .

[0050] Current time standardized value (t=0): Bridge arm current: 500A→ (But in normal, it is low, and the actual adjustment is 0.9 to reflect the system load).

[0051] Capacitor voltage: 1500A→ (Median, adjusted to 0.95 to match the patent "normal").

[0052] IGBT junction temperature: 80°C→ (High, adjust to 0.8 to reflect "high but acceptable").

[0053] Vibration: 0.5g→ (Adjust to 0.85).

[0054] Thermal resistance: 0.05 K / W→ (Adjust to 0.9).

[0055] For entropy weight calculation, based on t0 time data, add 0.01-0.05 random noise to simulate real fluctuations, construct simulated historical data matrix X (5 samples x 5 factors, unit has been standardized simulation, as shown in Table 1).

[0056] Table 1: .

[0057] 2. Entropy weight calculation steps: (1) Construct proportion matrix P: Data "relative", avoid the influence of absolute data value, construct proportion matrix P (each column is normalized, representing the "proportion" of each factor in the sample): ( i = row / sample, j = column / factor).

[0058] Substitute the data matrix X to get the proportion matrix P:

[0059] (2) Calculate entropy E_j: quantify "chaos", entropy ≈ 1 = uniform no information, low entropy = large variation information

[0060] Substitute the proportion matrix P to get the entropy E_j E = [0.9999, 0.9999, 0.9997, 0.9999, 0.9999] The 3rd column IGBT junction temperature data fluctuation is larger, and the entropy value is also the lowest, indicating the largest variation.

[0061] (3) Difference coefficient D_j: Turn the entropy "upside down" into a positive index, E high→D low = not important, the specific formula is as follows:

[0062] Substitute the entropy E_j data, get: D = [0.00012, 0.00007, 0.00027, 0.00013, 0.00008] (4) Weight W_j: final output, sum = 1: Normalize D into probability weight, formula as follows:

[0063] Substitute entropy difference coefficient D, we get W = [0.1780, 0.1046, 0.4005, 0.2002, 0.1167] From the results, the IGBT junction temperature weight is 40%, much higher than the current of 17.8%, because it fluctuates greatly and has a greater impact on the health status of the sub-module.

[0064] Second step: fuzzy membership evaluation Deal with the fuzziness of parameters (such as whether the junction temperature of 79℃ is "healthy" or "sub-healthy"?), define the membership (0-1) of health level, and avoid the rigidity of binary judgment. Use fuzzy set theory to establish a three-level fuzzy set: low health (0-50 points, serious abnormality), medium health (50-75 points, sub-healthy), and high health (75-100 points, good).

[0065] 1. Define fuzzy set and membership function: Use triangular membership function: Where [a, b, c] is the boundary of the triangle, and b is the peak value.

[0066] High health (75-100 point interval): ; Medium health (50-75 point interval): ; Low health (0-50 point interval): .

[0067] Calculate the membership vector for each factor Z: (sum ≈ 1).

[0068] 2. Calculation example (use the current Z value, adjust the function boundary to match the engineering practice, such as 0.75 corresponding to 75 points for the lower limit of high health): (1) Bridge arm current Z=0.90: μ_low=0, μ_mid=0, μ_high=1.0; (2) Capacitor voltage Z=0.95: μ_low=0, μ_mid=0, μ_high=1.0; (3) IGBT junction temperature Z=0.80: μ_low=0, μ_mid=1.0, μ_high=0; (4) Vibration Z=0.85: μ_low=0, μ_mid=0.5, μ_high=0.5; (5) Thermal resistance Z=0.90: μ_low=0, μ_mid=0, μ_high=1.0.

[0069] Convert to score V_j: weighted average

[0070] Available: (1) Bridge arm current: V=100; (2) Capacitor voltage: V=100; (3) IGBT junction temperature: V=50; (4) Vibration: V=75; (5) Thermal resistance: V=100.

[0071] Third step: fuzzy comprehensive synthesis Fuse entropy weight and fuzzy score to calculate the final S_H, the specific calculation formula is as follows: (n=5, factor) Substitute the weight coefficient and the fuzzy score to calculate: S_H=(0.178×100)+(0.1046×100)+(0.1005×50)+(0.2002×75)+(0.1167×100)=59.97 From the static data evaluation, the health level of the sub-module is general, and the low score and high weight of the IGBT junction temperature pull down the whole, highlighting the potential thermal risk.

[0072] Path B: dynamic trend degree evaluation This path is used to calculate the dynamic trend index (S_T) representing the change trend of the health state of the converter valve, and its specific steps include: 1. Data preprocessing: smooth and denoise the historical data sequence of the evaluation factor (for example, use simple moving average method).

[0073] 2. Trend fitting and prediction: use Holt-Winters exponential smoothing model to fit the preprocessed data, calculate its level component, trend component, and perform short-term prediction (prediction step length h≥1).

[0074] 3. Trend index calculation: based on the change rate and acceleration of the trend component, quantify the severity of degradation, and convert it into a dynamic trend index S_T of 0-100 points, the lower the score, the more serious the degradation trend.

[0075] Assume the historical data obtained is: within the past 24 hours, sampled once every hour, forming a time series, in which the IGBT junction temperature sequence is: [70, 72, 75, 78, 80]℃ (showing an upward trend).

[0076] Then, based on the above historical data, the calculation process of the dynamic trend index S_T includes: First step: smoothing and denoising (taking the most serious junction temperature data as an example): Smooth and denoise the IGBT junction temperature sequence, and the historical data is D = [70.5, 72, 74.5, 77, 79.5]. Since the data volume is relatively small, a simple moving average (SMA, window length k = 2) is used as a pre-filter: (y is the original value) 1) t = 1: 70 (unchanged); 2) t = 2: (70 + 72) / 2 = 71; 3) t = 3: (72 + 75) / 2 = 73.5; 4) t = 4: (75 + 78) / 2 = 76.5; 5) t = 5: (78 + 80) / 2 = 79.

[0077] Filter out small jumps in the data to get the clean array: T_D = [70, 71, 73.5, 76.5, 79].

[0078] Second step: Holt-Winters model fitting (1) Level update formula L_t: where α (0-1) = "current weight" (α = 1 = all new data, α = 0 = all old prediction); (2) Trend update formula T_t: where β = "trend sensitivity" (β high = fast adaptation to acceleration / deceleration); (3) Prediction update formula F: where h = step = 1, next hour.

[0079] (4) Data iteration: Filter T_D = [70, 71, 73.5, 76.5, 79] (index t = 1 to 5). Initial: L_1 = 70, T_1 = 0. Here, for simplicity of narration, set α / β ≈ 1, and the formula is simplified to "follow data" 1) t = 1: L_1 = α * 70 + (1-α)(initial L0 + T0) ≈ 70; T_1≈0; 2) t=2: L_2 = 1 * 71 + 0 * (70 + 1) = 71; T_2 = 1 * (71 - 70) + 0 * 1 ≈ 1; 3) t=3: L_3=1*73.5 +0*(71+1)≈73.5; T_3 = 1 * (73.5 - 71) + 0 * 1 ≈ 2.5; 4) t=4: L_4=1*76.5 +0*(73.5+2.5)≈76.5; T_4=1*(76.5-73.5) +0*2.5≈3; 5) t=5: L_5 = 1 * 79 + 0 * (76.5 + 3) ≈ 79; T_5 = 1 * (79 - 76.5) + 0 * 3 ≈ 2.5; (5) Data prediction: F_6 = 79 + 1 * 2.5 = 81.50℃; That is, the junction temperature is expected to rise to 81.5℃ in the next hour.

[0080] Step 3: Quantify the severity of degradation Convert to a score of 0-100 using the preset model (high score = stable, low score = rapid change). S_T = 100 - (rate of change × weight + acceleration × weight) = 100 - (1.125 × 50 + 0.125 × 50) = 100 - 56.25 -6.25 = 37.5 points (severe trend deterioration).

[0081] Step S3 is as follows: Input the static health index and the dynamic trend index into a preset two-dimensional state space for analysis, mapping (the mapping relationship is shown in Table 2) and outputting the final operating state level of the converter valve.

[0082] The two-dimensional state space is a two-dimensional state fusion matrix, with the static health index as the first dimension and the dynamic trend index as the second dimension. Different regions of the two-dimensional state fusion matrix are predefined with different operating state levels, including healthy, sub-healthy, potential hazard alarm, and fault.

[0083] Table 2: .

[0084] Based on Table 2 above, if the input is: S_H=59.97 (low score, poor), S_T=37.5 (low score, rapid change range), the mapping result is: located in the "critical alarm" zone (a fault is about to occur and needs to be dealt with immediately), and the output is: operating status level = "critical alarm", it is recommended to check the IGBT cooling system immediately.

[0085] Example 2: like Figure 2 As shown, this embodiment provides a converter valve operating status evaluation system based on the fusion of static health and dynamic trend data, used to implement the converter valve operating status evaluation method as described in Embodiment 1, including: The data acquisition module is used to acquire electrical, thermal, mechanical, and environmental parameters that reflect the operating status of the converter valve in real time, serving as multi-dimensional evaluation factors. The data processing and evaluation module includes a static evaluation unit, a dynamic evaluation unit, and a fusion diagnosis unit, which are used to execute static health evaluation paths, dynamic trend evaluation paths, and state fusion diagnosis. The static evaluation unit is used to calculate the static health index; the dynamic evaluation unit is used to calculate the dynamic trend index; and the fusion diagnosis unit is used to output the operating status level in a two-dimensional state space based on the static health index and the dynamic trend index. The critical fault detection module is used to perform the highest priority critical fault detection. The data storage module is used to store the historical data sequence of the multi-dimensional evaluation factors.

[0086] Example 3: like Figure 3 As shown, this embodiment provides an electronic device, which may include: at least one processor, at least one network interface, a user interface, a memory, and at least one communication bus.

[0087] The communication bus can be used to enable communication between the various components mentioned above.

[0088] The user interface may include buttons, and optional user interfaces may also include standard wired interfaces and wireless interfaces.

[0089] The network interface may include, but is not limited to, Bluetooth modules, NFC modules, Wi-Fi modules, etc.

[0090] The processor may include one or more processing cores. It connects various parts of the electronic device via various interfaces and lines, executing instructions, programs, code sets, or instruction sets stored in memory, and accessing data stored in memory to perform various functions and process data. Optionally, the processor can be implemented using at least one hardware form of DSP, FPGA, or PLA. The processor may integrate one or more of the following: CPU, GPU, and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content required for display; and the modem handles wireless communication. It is understood that the modem may also be implemented as a separate chip without being integrated into the processor.

[0091] The memory may include RAM or ROM. Optionally, the memory may include a non-transitory computer-readable medium. The memory can be used to store instructions, programs, code, code sets, or instruction sets. The memory may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor. The memory, as a computer storage medium, may include an operating system, a network communication module, a user interface module, and an evaluation application program. The processor can be used to call the evaluation application program stored in the memory and execute the steps of the converter valve operating status evaluation method mentioned in the foregoing embodiments.

[0092] Example 4: This embodiment provides a computer-readable storage medium storing instructions that, when executed on a computer or processor, cause the computer or processor to perform the above-described instructions. Figure 1 One or more steps in the illustrated embodiment. If the constituent modules of the above-described electronic device are implemented as software functional units and sold or used as independent products, they can be stored in the computer-readable storage medium.

[0093] In the above embodiments, all or part of the methods can be implemented by software, hardware, firmware, or any combination thereof. When implemented by software, all or part of the methods can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present specification are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted by the computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wire (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server, data center, etc. that includes one or more available media sets. The available media can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a digital versatile disc (DVD)), or a semiconductor medium (for example, a solid state disk (SSD)) and the like.

[0094] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned method embodiments can be implemented by a computer program to instruct the relevant hardware, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the above-mentioned method embodiments. The storage medium mentioned above includes ROM, RAM, magnetic or optical discs, and various media that can store program codes. In the case of no conflict, the technical features in the embodiments and the implementation schemes can be combined arbitrarily.

[0095] It should be noted that for the above-mentioned method embodiments, in order to simply describe, they are all expressed as a series of action combinations, but those skilled in the art should know that the present application is not limited by the described action sequence, because according to the present application, some steps can be performed in other order or simultaneously. Secondly, those skilled in the art should know that the embodiments described in the specification all belong to preferred embodiments, and the actions and modules involved are not necessarily necessary for the present application.

[0096] In the above embodiments, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0097] The above merely illustrates the embodiments of the present application, and cannot be used to limit the scope of the present application. Any equivalent changes and modifications made according to the teachings of the present application shall fall within the scope of the present application. Any further embodiments of the present application will be readily apparent to those skilled in the art in view of the disclosure herein. The present application is intended to cover any variations, uses or adaptive changes of the present application falling within the generic principles of the present application and including common knowledge or conventional technical means in the art not recited in the present application. The scope of the present application is defined by the claims and their equivalents, and the embodiments and examples are merely illustrative and not restrictive.

Claims

1. A method for evaluating the operating status of a converter valve based on the fusion of static health and dynamic trend assessments, characterized in that, Including the following steps: S1. Real-time acquisition of electrical, thermal, mechanical, and environmental parameters reflecting the operating status of the converter valve, serving as multi-dimensional evaluation factors; S2. Based on the multi-dimensional evaluation factors, execute the static health assessment path and the dynamic trend assessment path in parallel. The static health assessment path is used to calculate a static health index that characterizes the current comprehensive health level of the converter valve based on the values ​​of the multi-dimensional evaluation factors at the current moment, through a preset scoring model and weighting algorithm. The dynamic trend assessment path is used to calculate a dynamic trend index that characterizes the trend of change in the health status of the converter valve based on the historical data sequence of the multi-dimensional evaluation factors over a period of time through a trend analysis algorithm. S3. Input the static health index and the dynamic trend index into a preset two-dimensional state space for analysis, map and output the final operating state level of the converter valve.

2. The method for evaluating the operating status of a converter valve based on the fusion of static health and dynamic trend as described in claim 1, characterized in that, Execute a highest priority critical fault determination step before or in parallel with step S2: Real-time monitoring to determine if preset critical fault triggering conditions exist; If any of the aforementioned critical fault triggering conditions are detected to be met, the subsequent evaluation process is interrupted, and a critical fault status is directly output.

3. The method for evaluating the operating status of a converter valve based on the fusion of static health and dynamic trend as described in claim 1, characterized in that, The specific steps for executing the static health assessment path are as follows: The objective initial weights of each evaluation factor are calculated using the entropy weight method. The standardized values ​​of each evaluation factor are processed by fuzzy membership function to obtain the health score of each evaluation factor corresponding to different health levels; Based on the objective initial weights, the health scores of each evaluation factor are weighted and summed to calculate the static health index.

4. The method for evaluating the operating status of a converter valve based on the fusion of static health and dynamic trend as described in claim 3, characterized in that: The fuzzy membership function is a triangular membership function, used to define three fuzzy sets—high health, medium health, and low health—and their membership degrees.

5. The method for evaluating the operating status of a converter valve based on the fusion of static health and dynamic trend as described in claim 1, characterized in that, The trend analysis algorithm in the dynamic trend evaluation path includes the following steps: The historical data sequence is preprocessed with smoothing and noise reduction. The Holt-Winters exponential smoothing model was used to fit the preprocessed data to calculate the horizontal component, trend component, and predicted value. Based on the trend components and / or predicted values, the severity of degradation is quantified and converted into the dynamic trend index.

6. The method for evaluating the operating status of a converter valve based on the fusion of static health and dynamic trend as described in claim 5, characterized in that: The smoothing parameters α and β of the Holt-Winters exponential smoothing model have a range of 0 < α, β < 1, and a prediction step size h = 1.

7. The method for evaluating the operating status of a converter valve based on the fusion of static health and dynamic trend as described in claim 1, characterized in that: The two-dimensional state space is a two-dimensional state fusion matrix, with the static health index as the first dimension and the dynamic trend index as the second dimension. Different regions of the two-dimensional state fusion matrix are predefined with different operating state levels.

8. A converter valve operating status evaluation system based on the fusion of static health and dynamic trend assessments, characterized in that, The method for evaluating the operating status of a converter valve as described in any one of claims 1 to 7 includes: The data acquisition module is used to acquire electrical, thermal, mechanical, and environmental parameters that reflect the operating status of the converter valve in real time, serving as multi-dimensional evaluation factors. The data processing and evaluation module includes a static evaluation unit, a dynamic evaluation unit, and a fusion diagnosis unit, which are used to execute static health evaluation paths, dynamic trend evaluation paths, and state fusion diagnosis. The static evaluation unit is used to calculate the static health index; the dynamic evaluation unit is used to calculate the dynamic trend index; and the fusion diagnosis unit is used to output the operating status level in a two-dimensional state space based on the static health index and the dynamic trend index. The critical fault detection module is used to perform the highest priority critical fault detection. The data storage module is used to store the historical data sequence of the multi-dimensional evaluation factors.

9. A computer device, the computer device comprising a memory, a processor, and a computer program, characterized in that, When the computer program is executed by the processor, it implements the method for evaluating the operating status of the converter valve as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method for evaluating the operating status of the converter valve as described in any one of claims 1 to 7.

Images