Converter valve operation state prediction method and device, computer equipment and computer readable storage medium

Abstract

The invention relates to a converter valve operation state prediction method and device, computer equipment and a computer readable storage medium, and relates to the technical field of power grids. The method comprises the following steps: acquiring operation data of a converter valve in a preset time interval; based on a preset time interval, finely adjusting the pre-trained operation state prediction reference model to obtain an operation state prediction model in the preset time interval; based on an operation state prediction model in a preset time interval, performing feature extraction on the operation data to obtain operation features of the converter valve in the preset time interval; and obtaining an operation state prediction result of the converter valve in the preset time interval based on the operation state prediction model and the operation characteristics of the converter valve in the preset time interval. By adopting the method, the reliability of the predicted running state can be improved.

Description

Technical Field

[0001] The present application relates to the field of power grid technology, and in particular to a method, device, computer equipment, computer-readable storage medium and computer program product for predicting the operating state of a converter valve. Background Art

[0002] As a key component of the high-voltage direct current transmission system, the operating status of the converter valve is directly related to the safety, stability and operating efficiency of the entire system.

[0003] In the related art, the operating status of the converter valve is usually obtained by manually analyzing the operating data of the converter valve; however, the actual operation of the converter valve is easily affected by the dynamic influence of various external factors such as ambient temperature. Therefore, the reliability of the operating status obtained based on manual analysis is low. Summary of the invention

[0004] Based on this, it is necessary to provide a method, device, computer equipment and computer-readable storage medium for predicting the operating state of a converter valve, which can improve the reliability of the operating state, in order to address the technical problem of low reliability of the operating state.

[0005] In a first aspect, the present application provides a method for predicting the operating state of a converter valve, comprising:

[0006] Obtaining operation data of the converter valve in a preset time interval;

[0007] Based on the preset time interval, fine-tune the pre-trained operating state prediction benchmark model to obtain the operating state prediction model under the preset time interval;

[0008] Based on the operation state prediction model in the preset time interval, feature extraction is performed on the operation data to obtain the operation characteristics of the converter valve in the preset time interval;

[0009] Based on the operation state prediction model and the operation characteristics of the converter valve in the preset time interval, an operation state prediction result of the converter valve in the preset time interval is obtained.

[0010] In one embodiment, extracting features from the operating data to obtain operating features of the converter valve in the preset time interval includes:

[0011] Extracting features from the operating data to obtain candidate operating features of the converter valve in the preset time interval;

[0012] Determining the correlation between each candidate operating feature and the preset time interval;

[0013] Based on the correlation corresponding to each of the candidate operating characteristics, the operating characteristics of the converter valve in the preset time interval are screened out from the candidate operating characteristics.

[0014] In one of the embodiments, a plurality of infrared temperature sensors are provided in the operating environment of the converter valve, and the operating data includes the temperature collected by each infrared temperature sensor;

[0015] The extracting features from the operating data to obtain candidate operating features of the converter valve in the preset time interval includes:

[0016] Determine the number of multiple infrared temperature sensors;

[0017] Obtaining the ambient temperature of the operating environment based on the number of the plurality of infrared temperature sensors and the temperature collected by each infrared temperature sensor;

[0018] Obtaining an operating temperature of the converter valve based on a temperature regression coefficient corresponding to the preset time interval and the ambient temperature;

[0019] The operating temperature is determined as a candidate operating characteristic of the converter valve in the preset time interval.

[0020] In one of the embodiments, a plurality of ultraviolet sensors are provided in the operating environment of the converter valve, and the operating data includes ultraviolet values ​​collected by each ultraviolet sensor;

[0021] The extracting features from the operating data to obtain candidate operating features of the converter valve in the preset time interval includes:

[0022] Get the sensitivity of each UV sensor;

[0023] Based on the sensitivity of each ultraviolet sensor and the ultraviolet index collected by each ultraviolet sensor, obtaining the ultraviolet index in the operating environment;

[0024] The ultraviolet index is determined as a candidate operating characteristic of the converter valve in the preset time interval.

[0025] In one embodiment, the operation data includes the voltage and current of the converter valve when it is in operation;

[0026] The extracting features from the operating data to obtain candidate operating features of the converter valve in the preset time interval includes:

[0027] Determining the apparent power and the active power of the converter valve based on the voltage and the current;

[0028] Based on the apparent power, determining a load fluctuation index of the converter valve, and based on the apparent power and the active power, determining a power factor of the converter valve;

[0029] The load fluctuation index and the power factor are both determined as candidate operating characteristics of the converter valve in the preset time interval.

[0030] In one of the embodiments, the operation data includes a vibration signal of the converter valve during operation;

[0031] The extracting features from the operating data to obtain candidate operating features of the converter valve in the preset time interval includes:

[0032] Performing time domain analysis on the vibration signal to obtain an amplitude mean value and an amplitude root mean square value of the vibration signal;

[0033] Obtaining an amplitude index of the converter valve based on the amplitude mean value and the amplitude root mean square value;

[0034] The amplitude index is determined as a candidate operating characteristic of the converter valve in the preset time interval.

[0035] In one embodiment, obtaining the operation state prediction result of the converter valve in the preset time interval based on the operation state prediction model and the operation characteristics of the converter valve in the preset time interval includes:

[0036] Based on the operation state prediction model and the operation characteristics of the converter valve in the preset time interval, predicting the classification information of the converter valve in each preset operation state;

[0037] Based on the classification information of the converter valve in each preset operating state, at least one target operating state corresponding to the converter valve is determined in each preset operating state, and an operating state prediction result of the converter valve in the preset time interval is obtained.

[0038] In a second aspect, the present application further provides a device for predicting the operating state of a converter valve, comprising:

[0039] A data acquisition module, used to acquire the operation data of the converter valve in a preset time interval;

[0040] A model fine-tuning module, used to fine-tune the pre-trained operating state prediction benchmark model based on the preset time interval to obtain the operating state prediction model under the preset time interval;

[0041] a feature extraction module, configured to extract features from the operation data based on the operation state prediction model in the preset time interval, so as to obtain the operation features of the converter valve in the preset time interval;

[0042] The state prediction module is used to obtain the operation state prediction result of the converter valve in the preset time interval based on the operation state prediction model and the operation characteristics of the converter valve in the preset time interval.

[0043] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0044] Obtaining operation data of the converter valve in a preset time interval;

[0045] Based on the preset time interval, fine-tune the pre-trained operating state prediction benchmark model to obtain the operating state prediction model under the preset time interval;

[0046] Based on the operation state prediction model in the preset time interval, feature extraction is performed on the operation data to obtain the operation characteristics of the converter valve in the preset time interval;

[0047] Based on the operation state prediction model and the operation characteristics of the converter valve in the preset time interval, an operation state prediction result of the converter valve in the preset time interval is obtained.

[0048] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0049] Obtaining operation data of the converter valve in a preset time interval;

[0050] Based on the preset time interval, fine-tune the pre-trained operating state prediction benchmark model to obtain the operating state prediction model under the preset time interval;

[0051] Based on the operation state prediction model in the preset time interval, feature extraction is performed on the operation data to obtain the operation characteristics of the converter valve in the preset time interval;

[0052] Based on the operation state prediction model and the operation characteristics of the converter valve in the preset time interval, an operation state prediction result of the converter valve in the preset time interval is obtained.

[0053] The above-mentioned converter valve operation state prediction method, device, computer equipment and computer-readable storage medium first obtain the operation data of the converter valve in a preset time interval; then, based on the preset time interval, fine-tune the pre-trained operation state prediction benchmark model to obtain the operation state prediction model in the preset time interval; then, based on the operation state prediction model in the preset time interval, extract features from the operation data to obtain the operation characteristics of the converter valve in the preset time interval; finally, based on the operation state prediction model and the operation characteristics of the converter valve in the preset time interval, obtain the operation state prediction result of the converter valve in the preset time interval. In this way, by fine-tuning the operation state prediction benchmark model in the preset time interval, an operation state prediction model that is more suitable for the preset time interval can be obtained; based on the operation state prediction model that is more suitable for the preset time interval, feature extraction of the operation data and operation state prediction can be performed to obtain an operation state with higher reliability; the converter valve operation state prediction method based on the above process no longer relies on manual analysis, thereby improving the reliability of the operation state. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0055] FIG1 is a schematic flow chart of a method for predicting the operating state of a converter valve in one embodiment;

[0056] FIG2 is a flow chart showing the steps of extracting features from operating data to obtain operating features of a converter valve in a preset time interval in one embodiment;

[0057] FIG3 is a flow chart showing the steps of extracting features from operating data to obtain candidate operating features of a converter valve in a preset time interval in one embodiment;

[0058] FIG4 is a flow chart showing steps of extracting features from operating data to obtain candidate operating features of a converter valve in a preset time interval in another embodiment;

[0059] FIG5 is a flow chart showing the steps of extracting features from operating data to obtain candidate operating features of a converter valve in a preset time interval in another embodiment;

[0060] FIG6 is a flowchart of steps for extracting features from operating data to obtain candidate operating features of a converter valve in a preset time interval in yet another embodiment;

[0061] 7 is a flowchart of steps for obtaining an operating state prediction result of a converter valve in a preset time interval based on an operating state prediction model and operating characteristics of the converter valve in a preset time interval in one embodiment;

[0062] FIG8 is a structural block diagram of a device for predicting the operating state of a converter valve in one embodiment;

[0063] FIG. 9 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0065] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0066] In one embodiment, as shown in FIG1 , a method for predicting the operating state of a converter valve is provided. This embodiment uses the method applied to a server as an example for illustration. It is understandable that the method can also be applied to a terminal, and can also be applied to a system including a server and a terminal, and is implemented through the interaction between the server and the terminal; wherein the server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services; the terminal can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, etc. In this embodiment, the method includes the following steps:

[0067] Step S102, obtaining the operation data of the converter valve in a preset time interval.

[0068] Among them, the converter valve is the core component of the high-voltage direct current transmission system and plays an important role in AC-DC conversion.

[0069] Among them, various types of operating data of the converter valve include at least current, voltage, insulation resistance, power, temperature, switch action, vibration, sound, ultraviolet value, component data of various components, etc. when the converter valve is in operation.

[0070] Among them, the present application sets up multiple types of sensors at various key positions of the converter valve, and the sensing target of each type of sensor is a type of operating data; the server obtains the operating data of the converter valve through each sensor.

[0071] The operation data carries timestamp information of a preset time interval.

[0072] Specifically, the server obtains various operating data of the converter valve within a preset time interval based on various sensors that are set.

[0073] For each type of sensor, the server may be provided with at least one sensor.

[0074] In practical applications, operating data exists in the form of time series data.

[0075] Step S104, based on the preset time interval, fine-tune the pre-trained operating status prediction benchmark model to obtain the operating status prediction model in the preset time interval.

[0076] Among them, the operating status prediction benchmark model is a neural network model trained based on the historical operating data of the converter valve in different historical time intervals; in specific applications, the operating status prediction benchmark model is trained based on the LSTM (Long Short-Term Memory) neural network.

[0077] Specifically, the server inputs the operating data carrying timestamp information into the fine-tuned pre-trained operating status prediction benchmark model. The operating status prediction benchmark model determines the time attributes of the preset time interval based on the timestamp information, and fine-tunes its own model parameters based on the time attributes to obtain an operating status prediction model that is adapted to the time attributes of the preset time interval.

[0078] In practical applications, the operating status prediction benchmark model needs to back up a backup operating status prediction benchmark model, and fine-tune the model parameters based on the backup operating status prediction benchmark model to obtain an operating status prediction model with time attributes adapted to a preset time interval while retaining the operating status prediction benchmark model.

[0079] Among them, time attributes include season attributes and time period attributes; season attributes are used to characterize the season of the preset time interval in the four seasons of the year, such as spring, summer, autumn, and winter; time period attributes are used to characterize the time period of the preset time in a day, such as morning, noon, evening, early morning, etc.

[0080] In a specific application, when the preset time interval is short (e.g., a few hours), the operating status prediction benchmark model determines the seasonal attributes and time period attributes of the preset time interval based on the timestamp information of the preset time interval, and fine-tunes its own model parameters based on the seasonal attributes and time period attributes. When the preset time interval is long (e.g., more than 24 hours), the operating status prediction benchmark model determines the seasonal attributes of the preset time interval based on the timestamp information of the preset time interval, and fine-tunes its own model parameters based on the seasonal attributes.

[0081] In specific applications, when the server trains an operating status prediction benchmark model based on the historical operating data of the converter valve in different historical time intervals, in addition to training the operating status prediction benchmark model's ability to predict the operating status of the converter valve based on the historical operating data, it is also necessary to train the operating status prediction benchmark model's ability to fine-tune the model parameters based on the timestamp information of the historical operating data.

[0082] Step S106: Based on the operation status prediction model in the preset time interval, feature extraction is performed on the operation data to obtain the operation characteristics of the converter valve in the preset time interval.

[0083] Specifically, the operating state prediction benchmark model inputs the operating data into the operating state prediction model, and the operating state prediction model extracts features from the operating data to obtain a plurality of operating features of the converter valve in a preset time interval.

[0084] Step S108, obtaining an operating state prediction result of the converter valve in the preset time interval based on the operating state prediction model and the operating characteristics of the converter valve in the preset time interval.

[0085] Specifically, the operating state prediction model predicts the operating state of the converter valve in a preset time interval based on multiple operating characteristics of the converter valve in a preset time interval, and obtains a prediction result of the operating state of the converter valve in the preset time interval.

[0086] In the above-mentioned converter valve operation state prediction method, first, the server obtains the operation data of the converter valve in a preset time interval; then, based on the preset time interval, the server fine-tunes the pre-trained operation state prediction benchmark model to obtain the operation state prediction model in the preset time interval; then, based on the operation state prediction model in the preset time interval, the server extracts features from the operation data to obtain the operation characteristics of the converter valve in the preset time interval; finally, based on the operation state prediction model and the operation characteristics of the converter valve in the preset time interval, the server obtains the operation state prediction result of the converter valve in the preset time interval. In this way, by fine-tuning the operation state prediction benchmark model in the preset time interval, the server can obtain an operation state prediction model that is more suitable for the preset time interval; based on the operation state prediction model that is more suitable for the preset time interval, the server extracts features from the operation data and performs operation state prediction, and can obtain an operation state with higher reliability; the converter valve operation state prediction method based on the above process no longer relies on manual analysis, thereby improving the reliability of the operation state.

[0087] In an exemplary embodiment, as shown in FIG. 2 , the above step S106, extracting features from the operating data to obtain the operating features of the converter valve in a preset time interval, specifically includes the following steps:

[0088] Step S202: extracting features from the operation data to obtain candidate operation features of the converter valve in a preset time interval.

[0089] Step S204: determining the correlation between each candidate operation feature and a preset time interval.

[0090] Step S206: based on the correlation corresponding to each candidate operating feature, select the operating features of the converter valve in a preset time interval from the candidate operating features.

[0091] Specifically, the operation state prediction model first extracts features from the operation data to obtain multiple candidate operation features of the converter valve in a preset time interval. Since the actual operation of the converter valve is easily affected by the dynamic influence of multiple external factors such as ambient temperature, that is, the main influencing factors of the operation state of the converter valve are different in different preset time intervals, the server needs to predict the operation state of the converter valve based on different operation features in different preset time intervals. Then, the operation state prediction model determines the correlation between each candidate operation feature and the preset time interval, and based on the correlation corresponding to each candidate operation feature, selects the operation feature used for operation state prediction from multiple candidate operation features.

[0092] In a specific application, the operation status prediction model may use a principal component analysis method to determine the correlation between each candidate operation feature and a preset time interval, or may use a correlation coefficient method to determine the correlation between each candidate operation feature and a preset time interval.

[0093] In specific applications, the operation status prediction model can use the candidate operation features whose corresponding correlation is greater than a preset correlation threshold as the operation features for operation status prediction, or it can sort the candidate operation features in order of correlation from high to low, and select a preset number of candidate operation features with the top rankings as the operation features for status prediction.

[0094] In this embodiment, the server determines the operating characteristics corresponding to the preset time interval for operating status prediction through the operating status prediction model, so that the predicted operating status results can be more suitable for the preset time interval, further improving the reliability of the operating status.

[0095] In an exemplary embodiment, a plurality of infrared temperature sensors are provided in the operating environment of the converter valve, and the operating data includes the temperature collected by each infrared temperature sensor.

[0096] As shown in FIG. 3 , the above step S202, extracting features from the operation data to obtain candidate operation features of the converter valve in a preset time interval, specifically includes the following steps:

[0097] Step S302, determining the number of the plurality of infrared temperature sensors.

[0098] Step S304: obtaining the ambient temperature of the operating environment based on the number of the plurality of infrared temperature sensors and the temperature collected by each infrared temperature sensor.

[0099] Step S306, obtaining the operating temperature of the converter valve based on the temperature regression coefficient corresponding to the preset time interval and the ambient temperature.

[0100] Step S308: determining the operating temperature as a candidate operating characteristic of the converter valve in a preset time interval.

[0101] The present application sets a plurality of infrared temperature sensors in the operating environment of the converter valve. However, the temperature collected by the infrared temperature sensor does not represent the actual operating temperature of the converter valve, so the operating state prediction model also needs to calculate the actual operating temperature of the converter valve based on the temperature value collected by the infrared temperature sensor.

[0102] Specifically, the operation state prediction model first determines the number of multiple infrared temperature sensors, and then, as shown in Formula 1, the operation state prediction model calculates the ambient temperature of the operation environment in a preset time interval based on the number of multiple infrared temperature sensors and the temperature collected by each infrared temperature sensor:

[0103] (Formula 1)

[0104] in, is the ambient temperature of the operating environment; is the number of multiple infrared temperature sensors; For the The temperature value collected by an infrared temperature sensor.

[0105] Next, the operating state prediction model obtains the temperature regression coefficient corresponding to the preset time interval, and as shown in Formula 2, based on the temperature regression coefficient and the ambient temperature, calculates the operating temperature of the converter valve in the preset time interval:

[0106] (Formula 2)

[0107] in, is the operating temperature of the converter valve; and are all temperature regression coefficients corresponding to the preset time interval, which are used to perform regression fitting on the ambient temperature. In practical applications, the operating state prediction model is also pre-trained with a temperature regression fitting model, which is used to obtain the operating temperature of the converter valve based on the ambient temperature fitting of the operating environment of the converter valve under different time attributes.

[0108] Finally, the operation state prediction model determines the operation temperature of the converter valve as a candidate operation feature of the converter valve in a preset time interval.

[0109] In this embodiment, the operating state prediction model can further obtain the operating temperature of the converter valve based on the operating data of the temperature collected by the infrared temperature sensor, thereby truly reflecting the temperature condition of the converter valve during operation.

[0110] In an exemplary embodiment, a plurality of ultraviolet sensors are provided in the operating environment of the converter valve, and the operating data includes ultraviolet values ​​collected by each ultraviolet sensor.

[0111] The ultraviolet value collected by each ultraviolet sensor refers to the response value of the ultraviolet sensor.

[0112] As shown in FIG. 4 , the above step S202, extracting features from the operation data to obtain candidate operation features of the converter valve in a preset time interval, specifically includes the following steps:

[0113] Step S402, obtaining the sensitivity of each ultraviolet sensor.

[0114] Step S404: obtaining the ultraviolet index in the operating environment based on the sensitivity of each ultraviolet sensor and the ultraviolet index collected by each ultraviolet sensor.

[0115] Step S406: determining the ultraviolet index as a candidate operating characteristic of the converter valve in a preset time interval.

[0116] Among them, the UV index can characterize the discharge condition in the operating environment. Therefore, based on the UV index, the operating state prediction model can obtain the discharge condition of the converter valve during operation.

[0117] In specific applications, ultraviolet sensors are preferably arranged around converter valves and in areas where discharge may occur, such as near insulators.

[0118] The present application sets a plurality of ultraviolet sensors in the operating environment of the converter valve. Specifically, the operating state prediction model obtains the sensitivity of each ultraviolet sensor, and then, as shown in Formula 3, based on the sensitivity of each ultraviolet sensor and the ultraviolet index collected by each ultraviolet sensor, calculates the ultraviolet index of the operating environment in a preset time interval:

[0119] (Formula 3)

[0120] in, is the UV index in the operating environment; is the number of the multiple UV sensors; For the The sensitivity of each UV sensor; For the The UV values ​​collected by the UV sensors.

[0121] Finally, the operation status prediction model determines the UV index in the above operation environment as a candidate operation feature of the converter valve in a preset time interval.

[0122] In this embodiment, the operation status prediction model can further obtain the ultraviolet index in the operation environment based on the operation data of the ultraviolet value collected by the ultraviolet sensor, thereby truly reflecting the discharge condition of the converter valve during operation.

[0123] In an exemplary embodiment, the operation data includes voltage and current when the converter valve is in operation.

[0124] The voltage includes the voltage in each AC cycle in a preset time interval; and the current includes the current in each AC cycle in a preset time interval.

[0125] As shown in FIG. 5 , the above step S202, extracting features from the operation data to obtain candidate operation features of the converter valve in a preset time interval, specifically includes the following steps:

[0126] Step S502: Determine the apparent power and active power of the converter valve based on the voltage and current.

[0127] Step S504: determine the load fluctuation index of the converter valve based on the apparent power, and determine the power factor of the converter valve based on the apparent power and the active power.

[0128] Step S506: Determine the load fluctuation index and the power factor as candidate operating characteristics of the converter valve in a preset time interval.

[0129] Among them, the load fluctuation index is used to characterize the load fluctuation during the operation of the converter valve.

[0130] Specifically, first, as shown in Formula 4, for each AC cycle, the operation state prediction model obtains the apparent power of the converter valve in the AC cycle based on the voltage and current in the AC cycle:

[0131] (Formula 4)

[0132] in, is the apparent power of the converter valve; is the effective value of voltage; is the effective value of the current.

[0133] And, as shown in Formula 5, for each AC cycle, the operation state prediction model obtains the active power of the converter valve in the AC cycle based on the voltage and current in the AC cycle:

[0134] (Formula 5)

[0135] in, is the active power of the converter valve; is the length of the communication cycle; for Voltage value at the moment; for Current value at the moment.

[0136] Next, as shown in Formula 6, the operation state prediction model determines the load fluctuation index of the converter valve in a preset time interval based on the apparent power of the converter valve in each AC cycle:

[0137] (Formula 7)

[0138] in, is the load fluctuation index of the converter valve; is the maximum apparent power of the converter valve in each AC cycle; is the maximum apparent power of the converter valve in each AC cycle.

[0139] And, as shown in Formula 8, for each AC cycle, the operation state prediction model determines the power factor of the converter valve in the AC cycle based on the apparent power and active power in the AC cycle:

[0140] (Formula 8)

[0141] in, is the load fluctuation index of the converter valve; is the maximum apparent power of the converter valve in each AC cycle; is the minimum apparent power of the converter valve in each AC cycle.

[0142] Finally, the operation status prediction model determines the above load fluctuation index and power factor as candidate operation characteristics of the converter valve in a preset time interval.

[0143] In this embodiment, the operation status prediction model can further obtain the load fluctuation index and power factor of the converter valve based on the operation data such as voltage and current, so as to truly reflect the load fluctuation of the converter valve during operation and the impact of the load fluctuation on it.

[0144] In an exemplary embodiment, the operation data includes a vibration signal of the converter valve during operation.

[0145] As shown in FIG6 , the above step S202, extracting features from the operation data to obtain candidate operation features of the converter valve in a preset time interval, specifically includes the following steps:

[0146] Step S602: Perform time domain analysis on the vibration signal to obtain the amplitude mean and amplitude root mean square value of the vibration signal.

[0147] Step S604: obtaining an amplitude index of the converter valve based on the amplitude mean value and the amplitude root mean square value.

[0148] Step S606: determine the amplitude index as a candidate operating characteristic of the converter valve in a preset time interval.

[0149] The present application provides vibration sensors at key locations such as the reactor components of the converter valve. Specifically, the operation state prediction model obtains vibration signals collected by each vibration sensor.

[0150] Then, for each vibration signal, as shown in Formula 9, the operation status prediction model performs time domain analysis on the vibration signal to obtain the amplitude mean and amplitude root mean square value of the vibration signal:

[0151] (Formula 9)

[0152] in, is the RMS value of the amplitude of the vibration signal; is the mean amplitude of the vibration signal; The length of the preset time interval; for Amplitude value at the moment.

[0153] Then, as shown in Formula 10, the operation state prediction model obtains the amplitude index of the converter valve in a preset time interval based on the amplitude mean and the amplitude root mean square value of each vibration signal:

[0154] (Formula 10)

[0155] in, is the amplitude index of the converter valve; is the number of vibration signals; For the The root mean square value of the amplitude of a vibration signal; For the The mean amplitude of the vibration signal.

[0156] Finally, the operation status prediction model determines the amplitude index of the above-mentioned converter valve as a candidate operation feature of the converter valve in a preset time interval.

[0157] In this embodiment, the operation state prediction model can further obtain the amplitude index of the converter valve based on the operation data of the vibration signal, thereby truly reflecting the vibration condition of the converter valve during operation.

[0158] In an exemplary embodiment, the operation data of the converter valve also includes component data of each component.

[0159] Among them, each component includes at least a thyristor, a damping resistor and a capacitor. In a specific application, the component data of the thyristor includes the temperature and voltage of the thyristor, the component data of the damping resistor includes the temperature, voltage and resistance of the damping resistor, and the component data of the capacitor includes the temperature, voltage and capacitance of the capacitor.

[0160] In a specific application, the present application provides corresponding sensors at each component location to collect the above-mentioned component data of each component.

[0161] In an exemplary embodiment, in the above step 202, before extracting features from the operating data to obtain candidate operating features of the converter valve in a preset time interval, the following contents are specifically included: preprocessing the operating data to obtain preprocessed operating data.

[0162] The above step 202, extracting features from the operating data to obtain candidate operating features of the converter valve in a preset time interval, specifically also includes the following content: extracting features from the pre-processed operating data to obtain candidate operating features of the converter valve in a preset time interval.

[0163] Among them, data preprocessing at least includes data cleaning, outlier processing and normalization processing.

[0164] Among them, data cleaning refers to removing incomplete, inaccurate or irrelevant data from the original data to ensure the quality and consistency of the data set. It is specifically divided into missing value processing and redundant data removal.

[0165] Among them, outlier processing refers to identifying and processing abnormal data points in a data set; in specific applications, this application uses a box plot method to perform outlier processing.

[0166] Among them, normalization processing is to convert the data into a unified scale range to eliminate the differences between different dimensions.

[0167] In an exemplary embodiment, as shown in FIG. 7 , the above step S108, based on the operation state prediction model and the operation characteristics of the converter valve in the preset time interval, obtains the operation state prediction result of the converter valve in the preset time interval, specifically including the following steps:

[0168] Step S702: predicting classification information of the converter valve in each preset operating state based on the operating state prediction model and the operating characteristics of the converter valve in a preset time interval.

[0169] Step S704, based on the classification information of the converter valve in each preset operating state, at least one target operating state corresponding to the converter valve is determined in each preset operating state, and an operating state prediction result of the converter valve in a preset time interval is obtained.

[0170] The classification information is used to characterize the probability of the converter valve being in each preset operating state.

[0171] Among them, the preset operating states include normal operating state, abnormal temperature state (overheating or overcooling), abnormal vibration state (mechanical failure), abnormal power consumption state (overload or underload), key component failure state (such as thyristor failure), seasonal impact abnormal state, power fluctuation abnormal state, start-stop operation abnormal state, etc.

[0172] Specifically, the operation state prediction model calculates the classification probability of the converter valve corresponding to each preset operation state based on the various operation characteristics of the converter valve in the preset time interval, and obtains the classification information of the converter valve in each preset operation state; then, the operation state prediction model determines the preset operation state whose corresponding classification information is greater than the preset classification information threshold as the target operation state corresponding to the converter valve; finally, the operation state prediction model uses each target operation state of the converter valve as the operation state prediction result of the converter valve in the preset time interval and outputs it.

[0173] In this embodiment, the running state prediction is performed based on the running state prediction model that is more suitable for the preset time interval, so that a more reliable running state can be obtained, thereby improving the reliability of the running state.

[0174] In order to more clearly illustrate the method for predicting the operating state of a converter valve provided in the embodiment of the present application, the method for predicting the operating state of a converter valve is specifically described below with a specific embodiment, but it should be understood that the embodiment of the present application is not limited thereto. In one exemplary embodiment, the present application also provides a multi-scenario adaptive converter valve operating state intelligent monitoring method, which specifically includes the following steps:

[0175] 1. Data collection.

[0176] Sensors are used to collect various operating data of the converter valve in different seasons and different time periods of the same day.

[0177] 2. Input multiple types of operating data of the converter valve in different seasons and different time periods of the same day into the input layer of the LSTM model to be trained.

[0178] 3. The middle layer of the LSTM model processes multiple types of operating data of the converter valve in different seasons and different time periods of the same day, including data preprocessing and data feature extraction, to obtain multiple operating features of the converter valve in different seasons and different time periods of the same day;

[0179] 4. For each season and each time period, the LSTM model is trained based on multiple operating characteristics of the season and the time period, so that the LSTM model has the ability to predict the operating status of the converter valve based on the operating characteristics, as well as the ability to fine-tune the model parameters for different seasons and time periods, thereby obtaining a trained LSTM model.

[0180] 5. Intelligent monitoring of operating status.

[0181] Multiple types of operating data of the converter valve in a preset time interval are collected through sensors, and the multiple types of operating data are input into the input layer of the trained LSTM model. The trained LSTM model adaptively fine-tunes model parameters based on the preset time interval to better adapt to the preset time interval, and after fine-tuning, the operating status of the converter valve in the preset time interval is predicted based on the multiple types of operating data.

[0182] This embodiment can adaptively adjust the prediction results and state classification according to the actual data of the converter valve in different seasons and different operating conditions. Even for the same operating data, the trained LSTM model can dynamically identify and classify different operating states in the morning and evening or in different seasons, thereby improving the accuracy and reliability of monitoring and ensuring the safe and stable operation of the high-voltage direct current transmission system.

[0183] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0184] Based on the same inventive concept, the embodiment of the present application also provides a converter valve operation state prediction device for implementing the converter valve operation state prediction method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more converter valve operation state prediction device embodiments provided below can refer to the limitations of the converter valve operation state prediction method above, and will not be repeated here.

[0185] In an exemplary embodiment, as shown in FIG8 , a device for predicting the operating state of a converter valve is provided, comprising: a data acquisition module 802 , a model fine-tuning module 804 , a feature extraction module 806 and a state prediction module 808 , wherein:

[0186] The data acquisition module 802 is used to acquire the operation data of the converter valve in a preset time interval.

[0187] The model fine-tuning module 804 is used to fine-tune the pre-trained operating status prediction benchmark model based on a preset time interval to obtain an operating status prediction model under the preset time interval.

[0188] The feature extraction module 806 is used to extract features from the operation data based on the operation status prediction model in the preset time interval, so as to obtain the operation features of the converter valve in the preset time interval.

[0189] The state prediction module 808 is used to obtain the operation state prediction result of the converter valve in the preset time interval based on the operation state prediction model and the operation characteristics of the converter valve in the preset time interval.

[0190] In an exemplary embodiment, the feature extraction module 806 is also used to extract features from the operating data to obtain candidate operating features of the converter valve in a preset time interval; determine the correlation between each candidate operating feature and the preset time interval; and based on the correlation corresponding to each candidate operating feature, screen out the operating features of the converter valve in the preset time interval from the candidate operating features.

[0191] In an exemplary embodiment, a plurality of infrared temperature sensors are provided in the operating environment of the converter valve, and the operating data includes the temperature collected by each infrared temperature sensor.

[0192] The feature extraction module 806 is also used to determine the number of multiple infrared temperature sensors; obtain the ambient temperature of the operating environment based on the number of multiple infrared temperature sensors and the temperature collected by each infrared temperature sensor; obtain the operating temperature of the converter valve based on the temperature regression coefficient corresponding to the preset time interval and the ambient temperature; and determine the operating temperature as a candidate operating feature of the converter valve in the preset time interval.

[0193] In an exemplary embodiment, a plurality of ultraviolet sensors are provided in the operating environment of the converter valve, and the operating data includes ultraviolet values ​​collected by each ultraviolet sensor.

[0194] The feature extraction module 806 is also used to obtain the sensitivity of each ultraviolet sensor; based on the sensitivity of each ultraviolet sensor and the ultraviolet index collected by each ultraviolet sensor, obtain the ultraviolet index in the operating environment; and determine the ultraviolet index as a candidate operating feature of the converter valve in a preset time interval.

[0195] In an exemplary embodiment, the operation data includes voltage and current when the converter valve is in operation.

[0196] The feature extraction module 806 is also used to determine the apparent power and active power of the converter valve based on the voltage and current; determine the load fluctuation index of the converter valve based on the apparent power, and determine the power factor of the converter valve based on the apparent power and the active power; and determine the load fluctuation index and the power factor as candidate operating characteristics of the converter valve in a preset time interval.

[0197] In an exemplary embodiment, the operation data includes a vibration signal of the converter valve during operation.

[0198] The feature extraction module 806 is also used to perform time domain analysis on the vibration signal to obtain the amplitude mean and amplitude root mean square value of the vibration signal; based on the amplitude mean and amplitude root mean square value, obtain the amplitude index of the converter valve; and determine the amplitude index as a candidate operating feature of the converter valve in a preset time interval.

[0199] In an exemplary embodiment, the state prediction module 808 is also used to predict the classification information of the converter valve in each preset operating state based on the operating state prediction model and the operating characteristics of the converter valve in the preset time interval; based on the classification information of the converter valve in each preset operating state, at least one target operating state corresponding to the converter valve is determined in each preset operating state to obtain the operating state prediction result of the converter valve in the preset time interval.

[0200] Each module in the above-mentioned converter valve operation state prediction device can be implemented in whole or in part by software, hardware and their combination. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the corresponding operations of each of the above modules.

[0201] In an exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be shown in FIG9. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the operating data of the converter valve. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for predicting the operating state of the converter valve is implemented.

[0202] Those skilled in the art will understand that the structure shown in FIG. 9 is merely a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0203] In an exemplary embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0204] In an exemplary embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0205] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0206] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0207] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0208] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for predicting the operating state of a converter valve, characterized in that: The method comprises: Obtaining operation data of the converter valve in a preset time interval; Based on the preset time interval, fine-tune the pre-trained operating state prediction benchmark model to obtain the operating state prediction model under the preset time interval; Based on the operation state prediction model in the preset time interval, feature extraction is performed on the operation data to obtain the operation characteristics of the converter valve in the preset time interval; Based on the operation state prediction model and the operation characteristics of the converter valve in the preset time interval, an operation state prediction result of the converter valve in the preset time interval is obtained.

2. The method according to claim 1, characterized in that The extracting features of the operating data to obtain the operating features of the converter valve in the preset time interval includes: Extracting features from the operating data to obtain candidate operating features of the converter valve in the preset time interval; Determining the correlation between each candidate operating feature and the preset time interval; Based on the correlation corresponding to each of the candidate operating characteristics, the operating characteristics of the converter valve in the preset time interval are screened out from the candidate operating characteristics.

3. The method according to claim 2, characterized in that A plurality of infrared temperature sensors are provided in the operating environment of the converter valve, and the operating data includes the temperature collected by each infrared temperature sensor; The extracting features from the operating data to obtain candidate operating features of the converter valve in the preset time interval includes: Determine the number of multiple infrared temperature sensors; Obtaining the ambient temperature of the operating environment based on the number of the plurality of infrared temperature sensors and the temperature collected by each infrared temperature sensor; Obtaining an operating temperature of the converter valve based on a temperature regression coefficient corresponding to the preset time interval and the ambient temperature; The operating temperature is determined as a candidate operating characteristic of the converter valve in the preset time interval.

4. The method according to claim 2, characterized in that: A plurality of ultraviolet sensors are provided in the operating environment of the converter valve, and the operating data includes ultraviolet values ​​collected by each ultraviolet sensor; The extracting features from the operating data to obtain candidate operating features of the converter valve in the preset time interval includes: Get the sensitivity of each UV sensor; Based on the sensitivity of each ultraviolet sensor and the ultraviolet index collected by each ultraviolet sensor, obtaining the ultraviolet index in the operating environment; The ultraviolet index is determined as a candidate operating characteristic of the converter valve in the preset time interval.

5. The method according to claim 2, characterized in that: The operation data includes the voltage and current of the converter valve when it is in operation; The extracting features from the operating data to obtain candidate operating features of the converter valve in the preset time interval includes: Determining the apparent power and the active power of the converter valve based on the voltage and the current; Based on the apparent power, determining a load fluctuation index of the converter valve, and based on the apparent power and the active power, determining a power factor of the converter valve; The load fluctuation index and the power factor are both determined as candidate operating characteristics of the converter valve in the preset time interval.

6. The method according to claim 2, characterized in that The operation data includes a vibration signal of the converter valve when the converter valve is in operation; The extracting features from the operating data to obtain candidate operating features of the converter valve in the preset time interval includes: Performing time domain analysis on the vibration signal to obtain an amplitude mean value and an amplitude root mean square value of the vibration signal; Obtaining an amplitude index of the converter valve based on the amplitude mean value and the amplitude root mean square value; The amplitude index is determined as a candidate operating characteristic of the converter valve in the preset time interval.

7. The method according to any one of claims 1 to 6, characterized in that: The obtaining, based on the operation state prediction model and the operation characteristics of the converter valve in the preset time interval, an operation state prediction result of the converter valve in the preset time interval includes: Based on the operation state prediction model and the operation characteristics of the converter valve in the preset time interval, predicting the classification information of the converter valve in each preset operation state; Based on the classification information of the converter valve in each preset operating state, at least one target operating state corresponding to the converter valve is determined in each preset operating state, and an operating state prediction result of the converter valve in the preset time interval is obtained.

8. A device for predicting the operating state of a converter valve, characterized in that: The device comprises: A data acquisition module, used to acquire the operation data of the converter valve in a preset time interval; A model fine-tuning module, used to fine-tune the pre-trained operating state prediction benchmark model based on the preset time interval to obtain the operating state prediction model under the preset time interval; A feature extraction module, configured to extract features from the operation data based on the operation state prediction model in the preset time interval, so as to obtain the operation features of the converter valve in the preset time interval; The state prediction module is used to obtain the operation state prediction result of the converter valve in the preset time interval based on the operation state prediction model and the operation characteristics of the converter valve in the preset time interval.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images