PLC-based thermal power plant auxiliary control system and method

Abstract

The embodiment of the invention relates to the field of thermal power plants, and provides a thermal power plant auxiliary control system and method based on a PLC. A data receiving unit of the PLC is used for receiving the time queue of the rotating speed value of the steam turbine and the time queue of the environment temperature value; calculating the average value of the time queues of the turbine rotating speed values as a turbine rotating speed initial measurement value; determining a rotation speed measurement correction factor based on the time queue of ambient temperature values; multiplying the rotation speed measurement correction factor by the turbine rotation speed initial measurement value to obtain a turbine rotation speed optimization measurement value; and determining whether the rotating speed value of the steam turbine needs to be adjusted or not based on the comparison between the steam turbine rotating speed optimization measurement value and the steam turbine rotating speed safety threshold value. Therefore, the time sequence change of the environment temperature can be dynamically analyzed and quickly responded, and the steam turbine is ensured to always run in the optimal rotating speed interval, so that the measurement error caused by the quick change of the temperature is avoided.

Description

Technical Field

[0001] The present disclosure relates to the field of thermal power plants, and more specifically, to an auxiliary control system and method for a thermal power plant based on a PLC. Background Art

[0002] As an important part of the power system, the operating efficiency and safety of a thermal power plant are directly related to the stability and reliability of power supply. As one of the core equipment of a thermal power plant, the speed control of a steam turbine is crucial. Accurate speed measurement not only affects power generation efficiency but also has a decisive impact on the safe operation of the equipment. In actual operation, changes in ambient temperature can have a significant impact on the speed measurement of a steam turbine, resulting in measurement errors, which may in turn lead to unnecessary adjustments or failures.

[0003] Traditional temperature compensation methods mostly rely on fixed parameters or empirical formulas and are difficult to adapt to complex and variable actual working conditions. Especially under extreme temperature conditions, the effects of these methods are often poor. Specifically, traditional methods usually rely on fixed temperature compensation coefficients, which are determined under specific conditions (such as a laboratory environment). During the operation of a steam turbine, the ambient temperature may change rapidly, and fixed parameters cannot be adjusted in real time to cope with this dynamic change. Secondly, empirical formulas are usually derived based on historical data or theoretical models and are applicable to specific types of equipment or specific working conditions. When applied to different equipment or different working conditions, these formulas may no longer be effective.

[0004] Therefore, an optimized auxiliary control scheme for a thermal power plant is desired, which can correct the speed measurement with temperature compensation in a more intelligent way to adapt to complex and variable actual working conditions. Summary of the Invention

[0005] The present disclosure aims to solve at least one of the problems existing in the prior art and provides an auxiliary control system and method for a thermal power plant based on a PLC.

[0006] One aspect of the present disclosure provides an auxiliary control method for a thermal power plant based on a PLC, including:

[0007] Setting a safety threshold for the steam turbine speed in the PLC controller;

[0008] Using the data receiving unit of the PLC controller to receive the time queue of the steam turbine speed values collected by the speed sensor installed on the steam turbine;

[0009] Using the data receiving unit of the PLC controller to receive the time queue of the ambient temperature values collected by the temperature sensor installed near the speed sensor;

[0010] In the PLC controller, calculate the mean value of the time queue of the steam turbine speed value as the initial measured value of the steam turbine speed;

[0011] In the PLC controller, determine the speed measurement correction factor based on the time queue of the ambient temperature value;

[0012] In the PLC controller, multiply the speed measurement correction factor by the initial measured value of the steam turbine speed to obtain the optimized measured value of the steam turbine speed;

[0013] In the PLC controller, determine whether it is necessary to adjust the speed value of the steam turbine based on the comparison between the optimized measured value of the steam turbine speed and the safety threshold of the steam turbine speed.

[0014] Optionally, in the PLC controller, determining the speed measurement correction factor based on the time queue of the ambient temperature value includes:

[0015] Perform sequence equal division on the time queue of the ambient temperature value to obtain a set of ambient temperature time series subsequences;

[0016] Perform time series encoding based on the ambient temperature on each ambient temperature time series subsequence in the set of ambient temperature time series subsequences to obtain a set of ambient temperature local time series implicit feature vectors;

[0017] Perform conditional dependence type sequence feature enhancement on the set of ambient temperature local time series implicit feature vectors to obtain a set of ambient temperature local time series enhanced implicit feature vectors;

[0018] Obtain the speed measurement correction factor based on the set of ambient temperature local time series enhanced implicit feature vectors.

[0019] Optionally, performing time series encoding based on the ambient temperature on each ambient temperature time series subsequence in the set of ambient temperature time series subsequences to obtain a set of ambient temperature local time series implicit feature vectors includes: inputting each ambient temperature time series subsequence in the set of ambient temperature time series subsequences into a sequence encoder based on a forward LSTM model to obtain the set of ambient temperature local time series implicit feature vectors.

[0020] Optionally, performing conditional dependence type sequence feature enhancement on the set of ambient temperature local time series implicit feature vectors to obtain a set of ambient temperature local time series enhanced implicit feature vectors includes:

[0021] Perform field mapping on each ambient temperature local time series implicit feature vector in the set of ambient temperature local time series implicit feature vectors to obtain a set of incoming ambient temperature local time series implicit feature vectors;

[0022] Calculate the essential features of the set of local temporal implicit feature vectors of the incoming ambient temperature;

[0023] Based on the essential features, adaptively enhance each local temporal implicit feature vector of the set of local temporal implicit feature vectors of the ambient temperature to obtain the set of local temporal enhanced implicit feature vectors of the ambient temperature.

[0024] Optionally, calculating the essential features of the set of local temporal implicit feature vectors of the incoming ambient temperature includes:

[0025] Calculate the field depth factor of each local temporal implicit feature vector of the set of local temporal implicit feature vectors of the incoming ambient temperature to obtain the set of local temporal field depth factors of the ambient temperature;

[0026] Based on the set of local temporal field depth factors of the ambient temperature, calculate the ambient temperature temporal field essential feature vector of the set of local temporal implicit feature vectors of the incoming ambient temperature as the essential feature.

[0027] Optionally, based on the set of local temporal field depth factors of the ambient temperature, calculating the ambient temperature temporal field essential feature vector of the set of local temporal implicit feature vectors of the incoming ambient temperature includes:

[0028] Input the set of local temporal field depth factors of the ambient temperature into the softmax function for normalization processing to obtain the set of normalized local temporal field depth factors of the ambient temperature;

[0029] Using the set of normalized local temporal field depth factors of the ambient temperature as the set of weights, calculate the position-wise weighted sum of the set of local temporal implicit feature vectors of the incoming ambient temperature to obtain the ambient temperature temporal field essential feature vector.

[0030] Optionally, based on the essential features, adaptively enhancing each local temporal implicit feature vector of the set of local temporal implicit feature vectors of the ambient temperature to obtain the set of local temporal enhanced implicit feature vectors of the ambient temperature includes:

[0031] Using the ambient temperature temporal field essential feature vector as the conditional feature vector, perform fully connected encoding based on the Sigmoid function on each local temporal implicit feature vector of the set of local temporal implicit feature vectors of the ambient temperature to obtain the set of local temporal implicit fully connected encoded feature vectors;

[0032] Calculate the dot product of each pair of corresponding ambient temperature local temporal implicit feature vectors and ambient temperature local temporal implicit fully connected encoded feature vectors in the set of ambient temperature local temporal implicit feature vectors and the set of ambient temperature local temporal implicit fully connected encoded feature vectors to obtain the set of ambient temperature local temporal enhanced implicit feature vectors.

[0033] Optionally, obtaining the rotational speed measurement correction factor based on the set of ambient temperature local temporal enhanced implicit feature vectors includes:

[0034] Input each ambient temperature local temporal enhanced implicit feature vector in the set of ambient temperature local temporal enhanced implicit feature vectors into a correction factor generation module based on a decoder to obtain a set of local temperature correction factors;

[0035] Calculate the mean of the set of local temperature correction factors as the rotational speed measurement correction factor.

[0036] On the other hand, the present disclosure provides a PLC-based auxiliary control system for a thermal power plant, including:

[0037] A rotational speed safety threshold setting module for setting a steam turbine rotational speed safety threshold in the PLC controller;

[0038] A steam turbine rotational speed value acquisition module for using the data receiving unit of the PLC controller to receive a time queue of steam turbine rotational speed values collected by a rotational speed sensor installed on the steam turbine;

[0039] An ambient temperature value acquisition module for using the data receiving unit of the PLC controller to receive a time queue of ambient temperature values collected by a temperature sensor installed near the rotational speed sensor;

[0040] A steam turbine rotational speed initial measurement value calculation module for calculating the mean of the time queue of the steam turbine rotational speed values as the steam turbine rotational speed initial measurement value in the PLC controller;

[0041] A rotational speed measurement correction factor determination module for determining a rotational speed measurement correction factor based on the time queue of the ambient temperature values in the PLC controller;

[0042] A steam turbine rotational speed optimized measurement value calculation module for multiplying the rotational speed measurement correction factor by the steam turbine rotational speed initial measurement value to obtain a steam turbine rotational speed optimized measurement value in the PLC controller;

[0043] A comparison determination module for determining whether to adjust the rotational speed value of the steam turbine based on the comparison between the steam turbine rotational speed optimized measurement value and the steam turbine rotational speed safety threshold in the PLC controller.

[0044] Optionally, the rotational speed measurement correction factor determination module is configured to:

[0045] Sequentially dividing the time queue of the ambient temperature values into equal parts to obtain a set of ambient temperature time series subsequences;

[0046] Performing time series coding based on the ambient temperature on each ambient temperature time series subsequence in the set of the ambient temperature time series subsequences to obtain a set of ambient temperature local time series implicit feature vectors;

[0047] Performing condition-dependent sequence feature enhancement on the set of the ambient temperature local time series implicit feature vectors to obtain a set of ambient temperature local time series enhanced implicit feature vectors;

[0048] The rotation speed measurement correction factor is obtained based on the set of the ambient temperature local time series enhanced implicit feature vectors.

[0049] Compared with the prior art, the present disclosure uses machine learning data analysis and correction methods to perform sequence segmentation on the time queue of the ambient temperature values, and then performs local time series implicit encoding on each ambient temperature time series subsequence, and then automatically generates each local temperature correction factor based on the conditional dependency reinforcement representation of the implicit characteristics of each local time series of the ambient temperature, and calculates their average value to obtain the speed measurement correction factor. Unlike the traditional fixed parameter method, the present disclosure can dynamically analyze and quickly respond to the time series changes of the ambient temperature, ensuring that the turbine always operates within its optimal speed range, thereby avoiding speed measurement errors caused by rapid temperature changes. In addition, the present disclosure can automatically adjust according to different equipment and working conditions to improve the adaptability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] One or more embodiments are exemplarily illustrated by pictures in the corresponding drawings, and these exemplifications do not constitute limitations on the embodiments. Elements with the same reference numerals in the drawings represent similar elements, and unless otherwise stated, the figures in the drawings do not constitute proportional limitations.

[0051] Figure 1 is a schematic flow chart of a PLC-based auxiliary control method for a thermal power plant according to an embodiment of the present disclosure;

[0052] Figure 2 Schematic diagram of data flow of the PLC-based auxiliary control method for a thermal power plant according to an embodiment of the present disclosure;

[0053] Figure 3 Schematic flow chart of step S5 in the PLC-based auxiliary control method for a thermal power plant according to an embodiment of the present disclosure;

[0054] Figure 4 Schematic flowchart of step S53 in the PLC-based auxiliary control method for a thermal power plant according to an embodiment of the present disclosure;

[0055] Figure 5 Schematic flowchart of step S54 in the PLC-based auxiliary control method for a thermal power plant according to an embodiment of the present disclosure;

[0056] Figure 6 Schematic block diagram of the PLC-based auxiliary control system for a thermal power plant according to an embodiment of the present disclosure. Detailed implementation manners

[0057] To make the purposes, technical solutions and advantages of the embodiments of the present disclosure clearer, the embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. However, those of ordinary skill in the art can understand that, in the embodiments of the present disclosure, many technical details are presented for the convenience of readers to better understand the present disclosure. However, even without these technical details and various changes and modifications based on the following embodiments, the technical solutions claimed in the present disclosure can still be implemented. The following division of each embodiment is for convenience of description and should not constitute any limitation on the specific implementation manners of the present disclosure. Each embodiment can be combined and cross-referenced with each other on the premise of no contradiction.

[0058] Based on this, the present disclosure proposes a PLC-based auxiliary control method for a thermal power plant, which combines a PLC with an advanced data analysis algorithm and uses the ambient temperature to generate a more accurate rotational speed measurement correction factor to improve the intelligence level of the overall control system. Specifically, as Figure 1 and Figure 2 shown, the PLC-based auxiliary control method for a thermal power plant includes: S1, setting a steam turbine rotational speed safety threshold in the PLC controller; S2, using the data receiving unit of the PLC controller to receive the time queue of the steam turbine rotational speed values collected by a rotational speed sensor installed on the steam turbine; S3, using the data receiving unit of the PLC controller to receive the time queue of the ambient temperature values collected by a temperature sensor installed near the rotational speed sensor; S4, in the PLC controller, calculating the mean value of the time queue of the steam turbine rotational speed values as the initial steam turbine rotational speed measurement value; S5, in the PLC controller, determining a rotational speed measurement correction factor based on the time queue of the ambient temperature values; S6, in the PLC controller, multiplying the rotational speed measurement correction factor by the initial steam turbine rotational speed measurement value to obtain an optimized steam turbine rotational speed measurement value; S7, in the PLC controller, determining whether to adjust the rotational speed value of the steam turbine based on the comparison between the optimized steam turbine rotational speed measurement value and the steam turbine rotational speed safety threshold.

[0059] Exemplarily, in step S1, the steam turbine speed safety threshold is set in the PLC controller. It should be understood that one or a set of numerical values are predefined or configured inside the PLC in advance, and these numerical values represent the maximum and / or minimum speeds allowed for the steam turbine under normal and safe operating conditions. The steam turbine speed safety threshold is one of the key parameters to ensure the stable operation of the steam turbine and the entire power generation system. The setting of the steam turbine speed safety threshold is usually based on the technical specifications provided by the equipment manufacturer, the operating procedures of the power plant, and relevant industry standards. This threshold is used to prevent the steam turbine speed from being too high or too low, because abnormal speeds may cause mechanical damage, reduced efficiency, or even more serious safety accidents. For example, if the steam turbine speed exceeds its design limit, it may cause excessive wear of mechanical components, increased vibration, and even equipment failure; while too low speed may affect the quality and stability of power output.

[0060] When the steam turbine is running, the PLC continuously monitors the actual speed and compares it with the preset safety threshold. If the detected actual speed exceeds the set safety range, the PLC will trigger corresponding control actions, such as adjusting fuel supply, steam flow, or other control measures, to bring the steam turbine speed back within the safe range and ensure the safety and reliability of the system. For example, in a thermal power plant, the manufacturer of the steam turbine recommends that the optimal operating speed of this type of steam turbine is 3000 revolutions per minute (rpm). To ensure the safe operation of the equipment and leave a certain safety margin, the power plant engineer decides to set the safety threshold of the steam turbine speed in the PLC controller. Specifically, the maximum safe speed is set to 3100 rpm, and the minimum safe speed is set to 2900 rpm. This means that the actual operating speed of the steam turbine should always be maintained between these two values.

[0061] Exemplarily, in step S2, the data receiving unit of the PLC controller is used to receive the time queue of the steam turbine speed values collected by the speed sensor installed on the steam turbine. It should be understood that the PLC collects a series of speed data recorded by the speed sensor installed on the steam turbine through its built-in or connected data receiving unit. These data are arranged in chronological order to form a "time queue", that is, each data point is accompanied by a corresponding timestamp, indicating that the data was measured and recorded at a specific moment.

[0062] Specifically, the rotational speed sensor is usually installed at key positions of the steam turbine, such as the shaft end or other positions that can accurately reflect the change in the rotational speed of the steam turbine. When the steam turbine is operating, the sensor samples the rotational speed of the steam turbine at a fixed frequency (such as per second, per minute, etc.) and converts these sampled values into electrical signals or other forms of data for transmission to the PLC. To ensure the accuracy and real-time nature of data transmission, hardwired connections, wireless communication protocols (such as Wi-Fi, Zigbee), or industrial bus technologies (such as Modbus, Profibus) may be used between the sensor and the PLC for data exchange. The received data will be organized into a time series format within the PLC, which is the so-called "time queue". Such an organization method helps to analyze the time characteristics of the data during subsequent processing, such as calculating the average value, trend prediction, anomaly detection, etc. For a thermal power plant, it is crucial to maintain the steam turbine within a stable and safe operating rotational speed range, so an accurate and timely rotational speed monitoring mechanism is indispensable.

[0063] Exemplarily, in step S3, the data receiving unit of the PLC controller is used to receive the time queue of the ambient temperature values collected by the temperature sensor installed near the rotational speed sensor. It should be understood that the change in ambient temperature can significantly affect the performance of the steam turbine and the accuracy of the rotational speed sensor. For example, high temperature may cause materials to expand or contract, thereby affecting the gaps between mechanical components; low temperature may make the lubricating oil thicken, increasing the friction. In addition, temperature fluctuations may also cause drift of the sensor itself, resulting in inaccurate measurement results. Therefore, it is necessary to monitor the ambient temperature in real time and incorporate it into the control system. To achieve this, temperature sensors are installed near the steam turbine. Usually, these sensors are placed near the rotational speed sensor to more accurately capture the local temperature changes that have a direct impact on rotational speed measurement. The temperature sensor measures the ambient temperature at a certain frequency (such as per second, per minute, etc.) and sends the data, converted into electrical signals or other forms of data, to the data receiving unit of the PLC. The received temperature data is arranged in chronological order to form a "time queue". Each temperature data point is accompanied by a corresponding timestamp, indicating that the temperature was measured at a specific moment. Such time series data helps to analyze the temperature change trend during subsequent processing, such as calculating the average temperature, detecting abnormal temperature fluctuations, predicting future temperature changes, etc.

[0064] Exemplarily, in step S4, in the PLC controller, the mean value of the time queue of the steam turbine speed values is calculated as the initial measured value of the steam turbine speed. It should be understood that the speed sensor installed on the steam turbine will continuously send the actual speed information of the steam turbine to the data receiving unit of the PLC at a set time interval (such as per second, per minute, etc.). These data points form a time series, that is, a "time queue", and each data point is accompanied by a corresponding timestamp, indicating that the data is measured at a specific moment. This time-ordered data organization provides a basis for subsequent analysis. Due to various interference factors in the actual operating environment, such as mechanical vibration, instantaneous load changes, etc., the speed values measured once may fluctuate, and cannot fully reflect the true operating state of the steam turbine. By calculating the mean value of the speed values over a period of time, these short-term fluctuations can be smoothed out, so as to obtain a more stable and representative speed measurement value. This mean value is regarded as the "initial measured value of the steam turbine speed", which provides a benchmark for subsequent more accurate calibration and adjustment.

[0065] Exemplarily, in step S5, in the PLC controller, based on the time queue of the ambient temperature values, a speed measurement correction factor is determined. It should be understood that changes in the ambient temperature will cause problems such as thermal expansion and contraction of mechanical components, changes in material properties, and drift of sensor performance, all of which will affect the accuracy of speed measurement. Based on this, in the PLC controller, when determining the speed measurement correction factor based on the time queue of the ambient temperature values, the technical concept of the present disclosure is to use machine learning data analysis and correction methods to perform sequence segmentation on the time queue of the ambient temperature values, then perform local time series implicit encoding on each ambient temperature time series subsequence, and then automatically generate each local temperature correction factor according to the conditional dependence reinforcement representation of each ambient temperature local time series implicit feature, and calculate its mean value to obtain the speed measurement correction factor. Different from the traditional fixed parameter method, the present disclosure can dynamically analyze and quickly respond to the time series changes of the ambient temperature, ensure that the steam turbine always operates within its optimal speed range, thereby avoiding speed measurement errors caused by rapid temperature changes. Moreover, the present disclosure can automatically adjust according to different devices and working conditions to improve the adaptability of the system.

[0066] In one embodiment, as Figure 3As shown, in the PLC controller, determining a rotational speed measurement correction factor based on a time queue of the ambient temperature values includes: S51, performing sequence equal division on the time queue of the ambient temperature values to obtain a set of ambient temperature time series subsequences; S52, respectively performing time series encoding based on the ambient temperature on each of the ambient temperature time series subsequences in the set of ambient temperature time series subsequences to obtain a set of ambient temperature local time series implicit feature vectors; S53, performing conditional dependence type sequence feature enhancement on the set of ambient temperature local time series implicit feature vectors to obtain a set of ambient temperature local time series enhanced implicit feature vectors; S54, obtaining the rotational speed measurement correction factor based on the set of ambient temperature local time series enhanced implicit feature vectors.

[0067] Exemplarily, in step S51, performing sequence equal division on the time queue of the ambient temperature values to obtain a set of ambient temperature time series subsequences. It should be understood that considering that different time periods in the time queue of the ambient temperature values may have different time series characteristics or time series change trends, thus, in order to better capture the patterns and characteristics of the ambient temperature changes within each local time, in the technical solution of the present disclosure, performing sequence equal division on the time queue of the ambient temperature values to divide the data of the long sequence into smaller and manageable time series segments, obtaining a set of ambient temperature time series subsequences, providing rich data support for subsequent more fine-grained time series feature analysis.

[0068] In one embodiment, the time queue of the ambient temperature values is serially equally divided to obtain a set of ambient temperature time series subsequences, including: First, it is crucial to determine how to equally divide the original time queue. This depends on specific operating conditions requirements and technical considerations. For example, it can be divided at fixed time intervals (such as every minute, every hour), fixed number of samples (for example, each subsequence contains 100 data points), or according to the natural breakpoints of the temperature change trend. Selecting an appropriate equal division method can ensure that each subsequence can reflect sufficient information and will not be too long to lose local characteristics. Before actual operation, it is usually necessary to perform certain preprocessing on the original ambient temperature time queue. This includes identifying and excluding obviously incorrect or unreasonable temperature readings, that is, removing outliers; if there are data missing situations, appropriate methods need to be used for filling, such as linear interpolation method or other prediction algorithms; filters or other mathematical methods can also be used to reduce the influence of short-term fluctuations and make the temperature curve smoother. These preprocessing steps help improve the accuracy and reliability of subsequent analysis. Next, according to the preset principle, the preprocessed ambient temperature time queue is divided into several subsequences of equal length or with similar characteristics. Each subsequence represents the temperature change situation within a specific time period. For example, if the time interval method is adopted, the corresponding data segment can be intercepted according to the start time and end time; if it is based on the number of samples, a subsequence can be directly extracted from the original sequence every certain number of data points. This not only simplifies the subsequent computational complexity but also enables each subsequence to better reflect the main change trend of the temperature within that hour. Finally, all the divided subsequences are collected to form a "set of ambient temperature time series subsequences". Each element in this set is an independent small time series, and they together constitute a comprehensive description of the ambient temperature change during the entire monitoring period. These subsequences can be further used for advanced data analysis operations such as local time series implicit coding and conditional dependence type sequence feature enhancement, so as to provide a basis for finally generating an accurate rotational speed measurement correction factor.

[0069] Exemplarily, in step S52, each ambient temperature time series subsequence in the set of the ambient temperature time series subsequences is respectively subjected to time series encoding based on the ambient temperature to obtain a set of ambient temperature local time series implicit feature vectors. It should be understood that in order to further extract and capture the local temperature time series patterns in different ambient temperature time series subsequences, thereby effectively reflecting the trend and pattern of ambient temperature changes over time, including long-term dependencies, in one embodiment of the present disclosure, each ambient temperature time series subsequence in the set of the ambient temperature time series subsequences is respectively subjected to time series encoding based on the ambient temperature to obtain a set of ambient temperature local time series implicit feature vectors, including: inputting each ambient temperature time series subsequence in the set of the ambient temperature time series subsequences into a sequence encoder based on a forward LSTM model to obtain a set of ambient temperature local time series implicit feature vectors. It's worth noting that the forward LSTM model is a special type of recurrent neural network (RNN) designed to overcome the vanishing or exploding gradient problems encountered by traditional RNNs when processing long time series data. Specifically, "forward" refers to the direction of the data flow from the past to the future, meaning each element in the sequence is processed sequentially in chronological order. This directionality is well-suited for processing sequence data with a clear temporal order, such as in time series analysis. Therefore, by encoding the ambient temperature time series subsequences, the forward LSTM model can automatically extract implicit temperature features between different local times. These features not only reflect the raw temperature data but also carry complex information learned by the model, such as trends, periodicity, and anomalies. This is very helpful for subsequently generating more accurate correction factors.

[0070] Exemplarily, in step S53, the set of the local temporal implicit feature vectors of the ambient temperature is subjected to conditional-dependent sequence feature enhancement to obtain a set of local temporal enhanced implicit feature vectors of the ambient temperature. It should be understood that, considering that each local temporal implicit feature of the ambient temperature contains key and significant temporal information, and the contribution and importance of the ambient temperature features at different time points in the entire time domain are different. Therefore, in order to improve the ability of the features at each time point to distinguish different ambient temperature patterns, so as to extract deeper information of the ambient temperature at different time points, thereby providing a more refined and differentiated information representation for subsequent steps, in the technical solution disclosed in the present invention, the set of the local temporal implicit feature vectors of the ambient temperature is subjected to conditional-dependent sequence feature enhancement to obtain a set of local temporal enhanced implicit feature vectors of the ambient temperature. Through the conditional-dependent sequence feature enhancement mechanism, the conditional dependency between each ambient temperature change and the essential characteristics of the temperature within the entire time range can be dynamically considered, thereby accurately characterizing the ambient temperature characteristics at different moments.

[0071] In one embodiment, Figure 4As shown, conditional dependence-based sequence feature enhancement is performed on the set of local temporal implicit feature vectors of the ambient temperature to obtain a set of local temporal enhanced implicit feature vectors of the ambient temperature, including: S531, performing field mapping on each local temporal implicit feature vector in the set of local temporal implicit feature vectors of the ambient temperature to obtain a set of incoming ambient temperature local temporal implicit feature vectors; S532, calculating the essential features of the set of incoming ambient temperature local temporal implicit feature vectors; S533, based on the essential features, performing adaptive enhancement on each local temporal implicit feature vector in the set of local temporal implicit feature vectors of the ambient temperature to obtain the set of local temporal enhanced implicit feature vectors of the ambient temperature.

[0072] Exemplarily, in step S531, field mapping is performed on each local temporal implicit feature vector in the set of local temporal implicit feature vectors of the ambient temperature to obtain a set of incoming ambient temperature local temporal implicit feature vectors. Specifically, this process can be expressed by the formula:

[0073] h i = W1I i W2

[0074] where I i is the i-th local temporal implicit feature vector in the set of local temporal implicit feature vectors of the ambient temperature, W1 and W2 are the first weight matrix and the second weight matrix respectively, and h i is the incoming ambient temperature local temporal implicit feature vector corresponding to I i .

[0075] That is to say, "field mapping" can be understood as a feature transformation mechanism that maps each local temporal implicit feature vector of the ambient temperature to a new feature space - namely the "incoming environment". This new feature space is more focused on the requirements of the current task, that is, to find out how the ambient temperature change affects the turbine speed measurement. Through field mapping, the present disclosure can highlight those temperature change patterns closely related to the turbine speed measurement, while suppressing or ignoring less important information. The result of this is that the mapped feature vectors can more accurately reflect the impact of the ambient temperature on the turbine performance. In addition, field mapping helps to improve the discrimination ability of temperature features between different time points, so that even under similar temperature conditions, different temperature change trends and their impacts on the turbine can be clearly distinguished. This not only enhances the quality of feature expression, but also lays a solid foundation for generating more accurate speed measurement correction factors in the future.

[0076] In one embodiment, in step S532, calculating the essential features of the set of local temporal implicit feature vectors of the incoming ambient temperature includes: calculating the field depth factor of each local temporal implicit feature vector of the incoming ambient temperature in the set of local temporal implicit feature vectors of the incoming ambient temperature to obtain a set of local temporal field depth factors of the ambient temperature; based on the set of local temporal field depth factors of the ambient temperature, calculating the local temporal field essential feature vector of the set of local temporal implicit feature vectors of the incoming ambient temperature as the essential feature.

[0077] In one embodiment, based on the set of local temporal field depth factors of the ambient temperature, calculating the local temporal field essential feature vector of the set of local temporal implicit feature vectors of the incoming ambient temperature includes: inputting the set of local temporal field depth factors of the ambient temperature into the softmax function for normalization processing to obtain a set of normalized local temporal field depth factors of the ambient temperature; using the set of normalized local temporal field depth factors of the ambient temperature as a set of weights, calculating the position-wise weighted sum of the set of local temporal implicit feature vectors of the incoming ambient temperature to obtain the local temporal field essential feature vector. This process can be expressed by the formula:

[0078]

[0079] where, h i is the local temporal implicit feature vector of the incoming ambient temperature corresponding to I i , ‖·‖ 2 is the square of the calculated vector length, log2 represents the logarithmic function value with base 2, f(h i ) is the field depth factor for calculating h i , e i is the local temporal field depth factor of the ambient temperature corresponding to h i , exp(·) represents the calculation of the exponential function value with base e (the natural constant), a i is the normalized local temporal field depth factor of the ambient temperature corresponding to h i , and v c is the local temporal field essential feature vector of the ambient temperature.

[0080] It should be understood that the present disclosure has encoded the ambient temperature time series using a forward LSTM model, resulting in a set of implicit feature vectors for the local time series of the incoming ambient temperature. These feature vectors contain important information about temperature variations within a local time period, but their relative importance and impact may not be obvious. To better understand the role of each feature vector in the overall ambient temperature variation trend, the present disclosure calculates the "field depth factor" for each implicit feature vector for the local time series of the incoming ambient temperature. The field depth factor can be considered a metric that reflects the importance or complexity of each feature vector within its time period. By calculating the field depth factor for all implicit feature vectors for the local time series of the incoming ambient temperature, the present disclosure obtains a set of field depth factors for the local time series of the ambient temperature. Each factor in this set represents the degree of impact of temperature variations on turbine performance within the corresponding time period. For example, in certain time periods where temperature variations are drastic and frequent, the corresponding field depth factor will be higher; in contrast, in time periods where temperature variations are more stable, the field depth factor will be lower.

[0081] Based on these field depth factors, the next step is to calculate the essential eigenvector of the ambient temperature time series field, representing the set of implicit eigenvectors of the local time series of the incoming ambient temperature. This step aims to comprehensively consider all incoming eigenvectors and their corresponding field depth factors to extract a single eigenvector that comprehensively reflects the core attributes of ambient temperature variations throughout the entire monitoring period—the so-called "essential eigenvector of the ambient temperature time series field." This essential eigenvector not only condenses the key information from all incoming eigenvectors but also, through a weighted approach, highlights those temperature variation patterns that have a significant impact on turbine performance.

[0082] In one embodiment, in step S533, based on the essential features, each of the ambient temperature local time series implicit feature vectors in the set of the ambient temperature local time series implicit feature vectors is adaptively enhanced to obtain the set of the ambient temperature local time series enhanced implicit feature vectors, including: using the ambient temperature time series field essential feature vector as a conditional feature vector, performing full connection encoding based on the Sigmoid function on each of the ambient temperature local time series implicit feature vectors in the set of the ambient temperature local time series implicit feature vectors to obtain a set of ambient temperature local time series implicit fully connected encoded feature vectors; calculating the position point multiplication between each corresponding set of the ambient temperature local time series implicit feature vectors and the ambient temperature local time series implicit fully connected encoded feature vectors in the set of the ambient temperature local time series implicit feature vectors and the ambient temperature local time series implicit fully connected encoded feature vectors to obtain the set of the ambient temperature local time series enhanced implicit feature vectors. This process can be expressed by the formula:

[0083] O i= I i ⊙ [Sigmoid(M m v c + H m I i + b m )]

[0084] where I i is the i-th environmental temperature local temporal hidden feature vector in the set of environmental temperature local temporal hidden feature vectors, v c is the environmental temperature temporal field essential feature vector, M m is the environmental temperature temporal essential weight matrix, H m is the environmental temperature local temporal weight matrix, b m is the bias vector, Sigmoid(·) is the Sigmoid function, ⊙ is the element-wise multiplication, and O i is the environmental temperature local temporal enhanced hidden feature vector after temporal enhancement of I i .

[0085] That is, the essential feature vector of the environmental temperature time-series field calculated previously is used as the conditional feature vector. This essential feature vector condenses the key information of all the incoming feature vectors and highlights the temperature change patterns that have a significant impact on the steam turbine performance through a weighted method. Then, the present disclosure performs a fully connected encoding on each local time-series hidden feature vector of the environmental temperature using the Sigmoid function. The Sigmoid function is a commonly used activation function that can compress the input value between 0 and 1, making the output have the meaning of probability interpretation. This non-linear transformation helps to reveal the complex patterns hidden in the original data and provides a good basis for subsequent feature fusion. Next, for each local time-series hidden feature vector of the environmental temperature and its corresponding feature vector after being fully connected encoded by the Sigmoid function, the present disclosure calculates the element-wise multiplication between them. Element-wise multiplication means multiplying two values at the same position. This step effectively combines the information of the original feature vector and the encoded feature vector, forming a new feature representation, namely, the local time-series enhanced hidden feature vector of the environmental temperature. This method not only retains the basic information in the original feature vector but also adds the non-linear characteristics brought by the Sigmoid function encoding, enhancing the feature expression ability. Through such processing, each enhanced hidden feature vector becomes a comprehensive representation between the original feature vector and its encoded version. This comprehensive representation not only contains the time-series information of the original feature vector but also incorporates the deep features after non-linear transformation. Therefore, the set of local time-series enhanced hidden feature vectors of the environmental temperature can more accurately reflect the impact of environmental temperature changes on the steam turbine performance, especially when dealing with rapid changes or temperature fluctuations under extreme conditions, providing a more refined and reliable reference.

[0086] Exemplarily, in step S54, as Figure 5 shown, based on the set of the local time-series enhanced hidden feature vectors of the environmental temperature, obtaining the rotational speed measurement correction factor includes: S541, inputting each local time-series enhanced hidden feature vector in the set of the local time-series enhanced hidden feature vectors of the environmental temperature into a correction factor generation module based on a decoder to obtain a set of local temperature correction factors; S542, calculating the mean value of the set of the local temperature correction factors as the rotational speed measurement correction factor.

[0087] In step S541, each of the set of ambient temperature local temporal enhancement implicit feature vectors is input into a decoder-based correction factor generation module to obtain a set of local temperature correction factors. Specifically, the set of ambient temperature local temporal enhancement implicit feature vectors, obtained by conditionally enhancing the set of ambient temperature local temporal enhancement implicit feature vectors, is decoded to automatically generate the local temperature correction factors.

[0088] In one embodiment, a neural network-based decoder model is first initialized. This model can employ architectures such as a multi-layer perceptron (MLP), a recurrent neural network (RNN), or variations thereof such as a long short-term memory (LSTM) network. The decoder's task is to recover or predict the corresponding local temperature correction factor from the input enhanced latent feature vector. The model's input layer should match the dimensions of the enhanced latent feature vector, while the output layer produces a scalar value as the local temperature correction factor. Each local temporal enhanced latent feature vector of the ambient temperature is sequentially input into the decoder. Because these feature vectors have been enhanced, they carry rich information about the ambient temperature variation pattern, which helps the decoder more accurately predict the appropriate correction factor. Within the decoder, the input feature vector passes through a series of nonlinear transformation layers (e.g., fully connected layers and activation function layers). These layers are responsible for capturing the complex relationships in the feature vector and gradually transforming it into a representation that is closer to actual physical meaning. For example, using ReLU or Sigmoid activation functions can help reveal nonlinear characteristics hidden in the data; fully connected layers are used to enable interaction and combination between features. After forward propagation through the decoder, a value is generated at the output layer: the local temperature correction factor. This factor represents the impact of ambient temperature changes on turbine speed measurement during the current time period. It can be directly applied to turbine speed adjustment to compensate for measurement errors caused by temperature fluctuations. The above steps are repeated for each input ambient temperature local time series enhanced latent feature vector, ultimately resulting in a set of multiple local temperature correction factors.

[0089] In step S542, the mean of the set of local temperature correction factors is calculated as the speed measurement correction factor. Specifically, each local temperature correction factor represents the temperature correction within a local time segment. To more comprehensively characterize the speed correction value, the technical solution of the present disclosure calculates the mean of the set of local temperature correction factors as the speed measurement correction factor. In other words, the correction factor at a single time point may fluctuate due to transient temperature changes or noise. By calculating the mean of the local temperature correction factors, the correction information from all time points can be integrated to obtain a speed measurement correction factor that represents the overall ambient temperature impact. This helps ensure that the correction process takes into account temperature trends over the entire time period, thereby avoiding inaccurate corrections caused by individual outliers. This allows for dynamic analysis and rapid response to temporal changes in ambient temperature, ensuring that the turbine always operates within its optimal speed range, thereby avoiding measurement errors caused by rapid temperature changes. Furthermore, the present disclosure can automatically adjust to different equipment and operating conditions to improve system adaptability.

[0090] For example, in step S6, the PLC controller multiplies the speed measurement correction factor by the initial turbine speed measurement value to obtain an optimized turbine speed measurement value. It should be understood that in actual operating environments, turbine speed measurement is affected by a variety of factors, including but not limited to mechanical vibration, transient load changes, and most importantly, changes in ambient temperature. These factors may cause fluctuations in individual measurements and fail to fully reflect the actual operating status of the turbine. To overcome these issues, the system first calculates a representative "initial turbine speed measurement value," typically by averaging multiple speed data points collected over a period of time. However, this initial measurement value may still contain errors due to factors such as temperature fluctuations. To address these issues, the system further calculates a "speed measurement correction factor." This factor is extracted from the ambient temperature time series data through a series of complex algorithmic processing and reflects the specific impact of ambient temperature changes on turbine speed measurement. For example, if an increase in ambient temperature causes material expansion or sensor performance drift, the correction factor is adjusted accordingly to compensate for this effect. Therefore, the speed measurement correction factor is essentially a correction factor used to eliminate or reduce the effects of ambient temperature changes and other interfering factors on the speed measurement. Once the speed measurement correction factor and the initial turbine speed measurement are obtained, the next step is to perform a multiplication operation within the PLC controller. This multiplication uses the correction factor to fine-tune the initial measurement value, bringing the resulting "optimized turbine speed measurement" closer to the actual turbine operating speed. This not only improves measurement accuracy but also ensures that the turbine operates under optimal operating conditions, avoiding measurement errors caused by rapid temperature changes or other external factors.

[0091] For example, in step S7, the PLC controller determines whether the turbine speed value needs to be adjusted based on the comparison between the optimized measurement value of the turbine speed and the safety threshold of the turbine speed. It should be understood that the PLC continuously receives the optimized turbine speed measurement value from the sensor and compares the received optimized measurement value with the pre-set safety threshold. This can be achieved through a simple conditional statement (such as an if-else structure) to check whether the current speed exceeds the specified upper and lower limits. If it is detected that the optimized measurement value exceeds the maximum safe speed or is lower than the minimum safe speed, the PLC will trigger an alarm and start the corresponding control strategy to adjust the speed of the turbine. For example, when the speed is lower than the minimum safe value, the PLC may increase the fuel supply or increase the steam flow to increase the speed of the turbine; conversely, if the speed exceeds the maximum safe value, the PLC may reduce the fuel supply or adjust the load to slow down the speed of the turbine.

[0092] In addition, for situations that are close to but have not exceeded the safety threshold, the PLC can also adopt preventive measures to fine-tune the operating parameters of the steam turbine in advance to avoid potential problems. This forward-looking control method helps to maintain the stable operation of the system, reduce drastic adjustments in emergency situations, and thus extend the equipment life and improve the overall reliability.

[0093] In this way, the PLC can not only respond to abnormal situations in a timely manner but also provide more refined and stable control in daily operations. It can automatically adjust according to different working conditions without relying on fixed parameters or empirical formulas, thereby improving the adaptive ability of the system. Even in the face of complex and changeable actual working environments, such as extreme temperature changes or sudden load fluctuations, the PLC can ensure that the steam turbine always operates in the best and safe state.

[0094] In summary, the PLC-based auxiliary control method for thermal power plants according to the embodiments of the present disclosure is elucidated. It uses machine learning data analysis and correction methods to perform sequence segmentation on the time queue of the environmental temperature values, then performs local temporal implicit encoding on each environmental temperature time series subsequence, and then automatically generates each local temperature correction factor according to the conditional dependence reinforcement representation of each environmental temperature local temporal implicit feature, and calculates its mean value to obtain the rotational speed measurement correction factor. Different from the traditional fixed parameter method, the present disclosure can dynamically analyze and quickly respond to the temporal changes of the environmental temperature, ensure that the steam turbine always operates within its optimal rotational speed range, and thus avoid the rotational speed measurement error caused by rapid temperature changes. Moreover, the present disclosure can automatically adjust according to different equipment and working conditions to improve the adaptability of the system.

[0095] Figure 6 Schematic block diagram of the PLC-based auxiliary control system for thermal power plants according to the embodiments of the present disclosure. As Figure 6As shown, the PLC-based auxiliary control system 100 of a thermal power plant includes: a rotational speed safety threshold setting module 110 for setting the safety threshold of the steam turbine rotational speed in the PLC controller; a steam turbine rotational speed value acquisition module 120 for using the data receiving unit of the PLC controller to receive the time queue of the steam turbine rotational speed values collected by a rotational speed sensor installed on the steam turbine; an ambient temperature value acquisition module 130 for using the data receiving unit of the PLC controller to receive the time queue of the ambient temperature values collected by a temperature sensor installed near the rotational speed sensor; a steam turbine rotational speed initial measurement value calculation module 140 for calculating, in the PLC controller, the mean value of the time queue of the steam turbine rotational speed values as the initial measurement value of the steam turbine rotational speed; a rotational speed measurement correction factor determination module 150 for determining, in the PLC controller, a rotational speed measurement correction factor based on the time queue of the ambient temperature values; a steam turbine rotational speed optimized measurement value calculation module 160 for multiplying, in the PLC controller, the rotational speed measurement correction factor by the initial measurement value of the steam turbine rotational speed to obtain the optimized measurement value of the steam turbine rotational speed; and a comparison and determination module 170 for determining, in the PLC controller, whether it is necessary to adjust the rotational speed value of the steam turbine based on the comparison between the optimized measurement value of the steam turbine rotational speed and the safety threshold of the steam turbine rotational speed.

[0096] In one embodiment, the rotational speed measurement correction factor determination module is configured to: equally divide the time queue of the ambient temperature values to obtain a set of ambient temperature time series subsequences; perform time series encoding based on the ambient temperature on each ambient temperature time series subsequence in the set of ambient temperature time series subsequences to obtain a set of ambient temperature local time series implicit feature vectors; perform conditional dependence type sequence feature enhancement on the set of ambient temperature local time series implicit feature vectors to obtain a set of ambient temperature local time series enhanced implicit feature vectors; and obtain the rotational speed measurement correction factor based on the set of ambient temperature local time series enhanced implicit feature vectors.

[0097] Here, those skilled in the art can understand that the specific operations of the above-mentioned various modules and units in the PLC-based auxiliary control system of a thermal power plant have been described in detail in the description of the Figures 1 to 5 PLC-based auxiliary control method of a thermal power plant, and thus, the repeated description thereof will be omitted.

[0098] Those of ordinary skill in the art can understand that the above-mentioned various embodiments are specific embodiments for implementing the present disclosure, and in practical applications, various changes can be made in form and details without departing from the spirit and scope of the present disclosure.

Claims

1. A PLC-based auxiliary control method for thermal power plants, characterized in that Including: Set the steam turbine speed safety threshold in the PLC controller; Use the data receiving unit of the PLC controller to receive the time queue of the steam turbine speed values collected by the speed sensors installed on the steam turbine; Use the data receiving unit of the PLC controller to receive the time queue of the ambient temperature values collected by the temperature sensors installed near the speed sensors; In the PLC controller, calculate the mean value of the time queue of the steam turbine speed values as the initial measured value of the steam turbine speed; In the PLC controller, determine the speed measurement correction factor based on the time queue of the ambient temperature values; In the PLC controller, multiply the speed measurement correction factor by the initial measured value of the steam turbine speed to obtain the optimized measured value of the steam turbine speed; In the PLC controller, determine whether it is necessary to adjust the speed value of the steam turbine based on the comparison between the optimized measured value of the steam turbine speed and the steam turbine speed safety threshold; 2. The PLC-based auxiliary control method for a thermal power plant according to claim 1, wherein In the PLC controller, determine the speed measurement correction factor based on the time queue of the ambient temperature values, including: Perform sequence equal division on the time queue of the ambient temperature values to obtain a set of ambient temperature time series subsequences; Perform time series encoding based on the ambient temperature on each ambient temperature time series subsequence in the set of ambient temperature time series subsequences to obtain a set of ambient temperature local time series implicit feature vectors; Perform conditional dependence type sequence feature enhancement on the set of ambient temperature local time series implicit feature vectors to obtain a set of ambient temperature local time series enhanced implicit feature vectors; Based on the set of ambient temperature local time series enhanced implicit feature vectors, obtain the speed measurement correction factor; 3. The PLC-based auxiliary control method for a thermal power plant according to claim 2, wherein Perform time series encoding based on the ambient temperature on each ambient temperature time series subsequence in the set of ambient temperature time series subsequences to obtain a set of ambient temperature local time series implicit feature vectors, including: inputting each ambient temperature time series subsequence in the set of ambient temperature time series subsequences into the sequence encoder based on the forward LSTM model to obtain the set of ambient temperature local time series implicit feature vectors; 4. The PLC-based auxiliary control method for thermal power plants according to claim 3, characterized in that Perform conditional dependence type sequence feature enhancement on the set of ambient temperature local time series implicit feature vectors to obtain a set of ambient temperature local time series enhanced implicit feature vectors, including: Perform field mapping on each ambient temperature local time series implicit feature vector in the set of ambient temperature local time series implicit feature vectors to obtain a set of incoming ambient temperature local time series implicit feature vectors; Calculate the essential features of the set of incoming ambient temperature local time series implicit feature vectors; Based on the essential features, perform adaptive enhancement on each ambient temperature local time series implicit feature vector in the set of ambient temperature local time series implicit feature vectors to obtain the set of ambient temperature local time series enhanced implicit feature vectors; 5. The PLC-based auxiliary control method for a thermal power plant according to claim 4, wherein Calculate the essential features of the set of incoming ambient temperature local time series implicit feature vectors, including: Calculate the field depth factor of each local temporal implicit feature vector of the ambient temperature at the entrance in the set of local temporal implicit feature vectors of the ambient temperature at the entrance to obtain a set of field depth factors of the local temporal field of the ambient temperature; Based on the set of field depth factors of the local temporal field of the ambient temperature, calculate the essential feature vector of the temporal field of the ambient temperature of the set of local temporal implicit feature vectors of the ambient temperature at the entrance as the essential feature.

6. The PLC-based auxiliary control method for a thermal power plant according to claim 5, wherein, Based on the set of field depth factors of the local temporal field of the ambient temperature, calculating the essential feature vector of the temporal field of the ambient temperature of the set of local temporal implicit feature vectors of the ambient temperature at the entrance includes: Input the set of field depth factors of the local temporal field of the ambient temperature into the softmax function for normalization processing to obtain a set of normalized field depth factors of the local temporal field of the ambient temperature; Taking the set of normalized field depth factors of the local temporal field of the ambient temperature as a set of weights, calculate the position-wise weighted sum of the set of local temporal implicit feature vectors of the ambient temperature at the entrance to obtain the essential feature vector of the temporal field of the ambient temperature.

7. The PLC-based auxiliary control method for a thermal power plant according to claim 6, wherein, Based on the essential feature, adaptively enhance each local temporal implicit feature vector of the ambient temperature in the set of local temporal implicit feature vectors of the ambient temperature to obtain a set of locally temporally enhanced implicit feature vectors of the ambient temperature, including: Using the essential feature vector of the temporal field of the ambient temperature as a conditional feature vector, perform fully connected encoding based on the Sigmoid function on each local temporal implicit feature vector of the ambient temperature in the set of local temporal implicit feature vectors of the ambient temperature to obtain a set of locally temporally implicit fully connected encoded feature vectors of the ambient temperature; Calculate the position-wise dot product between each pair of corresponding local temporal implicit feature vectors and locally temporally implicit fully connected encoded feature vectors in the set of local temporal implicit feature vectors of the ambient temperature and the set of locally temporally implicit fully connected encoded feature vectors of the ambient temperature to obtain a set of locally temporally enhanced implicit feature vectors of the ambient temperature.

8. The PLC-based auxiliary control method for a thermal power plant according to claim 7, wherein, Based on the set of locally temporally enhanced implicit feature vectors of the ambient temperature, obtain the rotational speed measurement correction factor, including: Input each locally temporally enhanced implicit feature vector of the ambient temperature in the set of locally temporally enhanced implicit feature vectors of the ambient temperature into a correction factor generation module based on a decoder to obtain a set of local temperature correction factors; Calculate the mean of the set of local temperature correction factors as the rotational speed measurement correction factor.

9. A PLC-based auxiliary control system for a thermal power plant, characterized in that, Including: A rotational speed safety threshold setting module for setting a steam turbine rotational speed safety threshold in the PLC controller; A steam turbine rotational speed value acquisition module for using the data receiving unit of the PLC controller to receive a time queue of steam turbine rotational speed values collected by a rotational speed sensor installed on the steam turbine; An ambient temperature value acquisition module for using the data receiving unit of the PLC controller to receive a time queue of ambient temperature values collected by a temperature sensor installed near the rotational speed sensor; The initial turbine speed measurement value calculation module is used to calculate the mean value of the time queue of the turbine speed value in the PLC controller as the initial turbine speed measurement value; The rotation speed measurement correction factor determination module is used to determine the rotation speed measurement correction factor based on the time queue of the ambient temperature value in the PLC controller; The optimized turbine speed measurement value calculation module is used to multiply the rotation speed measurement correction factor by the initial turbine speed measurement value in the PLC controller to obtain the optimized turbine speed measurement value; The comparison and determination module is used to determine whether it is necessary to adjust the rotation speed value of the turbine based on the comparison between the optimized turbine speed measurement value and the turbine speed safety threshold in the PLC controller.

10. The auxiliary control system for a thermal power plant based on a PLC according to claim 9, characterized in that, The rotation speed measurement correction factor determination module is used for: Sequentially dividing the time queue of the ambient temperature value to obtain a set of ambient temperature time series subsequences; Performing time series encoding based on the ambient temperature on each ambient temperature time series subsequence in the set of ambient temperature time series subsequences to obtain a set of ambient temperature local time series implicit feature vectors; Performing conditional dependence type sequence feature enhancement on the set of ambient temperature local time series implicit feature vectors to obtain a set of ambient temperature local time series enhanced implicit feature vectors; Obtaining the rotation speed measurement correction factor based on the set of ambient temperature local time series enhanced implicit feature vectors.

Images