Control system and method of power station air cooling device, electronic equipment and storage medium

Abstract

The invention relates to the field of air cooling, and provides a control system and method for a power station air cooling device, electronic equipment and a storage medium, and the system comprises a plurality of sensors which are used for collecting different types of performance parameters of the power station air cooling device; the server is used for determining the refrigeration efficiency of the power station air cooling device based on the performance parameters and optimizing the power station air cooling device when the refrigeration efficiency is lower than a set threshold value; the method for optimizing the power station air cooling device comprises the step of controlling operation parameters of the power station air cooling device based on an optimization control model obtained through pre-training and a preset performance parameter target value. The power station air cooling device is used for overcoming the defects that an existing power station air cooling system is low in cooling efficiency, high in energy consumption and difficult to achieve the optimal operation effect, performance parameters of the power station air cooling device can be collected in real time, the power station air cooling device is optimized based on data collected in real time, and the refrigeration efficiency of the power station air cooling device can be effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of air cooling technology, and in particular to a control system and method, electronic equipment and storage medium for an air cooling device of a power station. Background Art

[0002] With the continuous development of the power industry, the performance of the power plant air cooling system, as an important cooling device, has a crucial impact on the overall operation of the power plant.

[0003] The quality of the power plant air cooling system is mainly evaluated by the cooling efficiency. The cooling efficiency, that is, the energy consumption required to achieve the same cooling efficiency, is different. For the power plant air cooling system, it usually controls the cooling effect through a variety of operating parameters. The operating parameters may include fan speed and shutter opening, etc. Different operating parameters have different advantages and disadvantages when cooling. For example, different operating parameters may have different cooling assistance and energy consumption.

[0004] At present, when adjusting the cooling efficiency, the traditional power plant air cooling system control mainly relies on experience to perform simple control of various types of operating parameters. It is unable to accurately adapt to complex and changeable working conditions and environmental conditions in real time, resulting in low cooling efficiency in some operating periods, inability to fully utilize equipment performance, and high system energy consumption, which increases the operating costs of the power plant and makes it difficult to achieve the overall optimal operating effect. Summary of the invention

[0005] The present invention provides a control system and method, electronic equipment, and storage medium for a power plant air-cooling device, which are used to solve the defects of the existing power plant air-cooling system, such as low cooling efficiency and high energy consumption, and difficulty in achieving optimal operating effects. The solution of the present application can collect the performance parameters of the power plant air-cooling device in real time, and optimize the power plant air-cooling device based on the real-time collected data, which can effectively improve the refrigeration efficiency of the power plant air-cooling device.

[0006] The present invention provides a control system for a power station air cooling device, comprising:

[0007] Several sensors are used to collect different types of performance parameters of the power plant air cooling device;

[0008] A server, configured to determine a cooling efficiency of an air cooling device of a power plant based on performance parameters, and to optimize the air cooling device of the power plant when the cooling efficiency is lower than a set threshold;

[0009] Methods for optimizing power plant air cooling devices include:

[0010] The operating parameters of the power plant air cooling device are controlled based on the pre-trained optimization control model and the preset target values ​​of the performance parameters. The training set of the optimization control model is the first training set. The first training set includes the operating parameters and performance parameters of the power plant air cooling device under different working conditions. The operating parameters are the parameters used by the power plant air cooling device to control the cooling effect.

[0011] According to the control system of the power plant air cooling device provided by the present invention, the optimization control model is trained by the following method:

[0012] Fuzzify the data in the first training set;

[0013] After fuzzy reasoning is performed on the performance parameters and the operating parameters, the data in the first training set is input into the optimization control initial model, and the parameters of the optimization control initial model are adjusted to determine the operating parameters corresponding to the performance parameters in the first training set;

[0014] After iteratively optimizing the performance parameters and operating parameters in the initial optimization control model, the optimization control model is output.

[0015] According to the control system of the power plant air cooling device provided by the present invention, the iterative optimization process of the optimization control model includes:

[0016] Performing fuzzy processing on the data in the first training set, where the first training set includes performance parameters and operating parameters of the power plant air cooling device under different operating conditions;

[0017] Inputting the data in the first training set into the optimization control initial model to determine the operating parameters corresponding to the performance parameters in the first training set;

[0018] Compare the operating parameters output by the initial optimization control model with the pre-determined optimal operating parameters to determine the error value. The optimal operating parameters enable the power plant air cooling device to achieve the optimal cooling efficiency.

[0019] If the error value exceeds a preset error threshold, the fuzzy processing process is reversely adjusted based on the error value, and the error value is re-determined.

[0020] According to the control system of the power plant air cooling device provided by the present invention, the server is also used for:

[0021] Extract features of performance parameters collected by sensors based on the pre-trained fault prediction model;

[0022] The performance parameters after feature extraction are subjected to feature fusion and feature abstraction to obtain fusion features, which reflect the operating fault conditions of the air cooling device of the power station;

[0023] Based on the fused features, a fault detection result of the power plant air cooling device is output, and the fault detection result at least characterizes the fault type of the power plant air cooling device.

[0024] According to the control system of the power plant air cooling device provided by the present invention, the training process of the fault prediction model includes:

[0025] Based on the convolutional neural network model and the recurrent neural network model, an initial fault detection model is constructed;

[0026] After initializing the parameters of the initial fault detection model, the data in the second training set is input into the initial fault detection model for model training. The data in the second training set includes performance parameters of the power plant air cooling device under different fault conditions and corresponding fault labels;

[0027] Obtain the output result of the initial fault detection model, and calculate the loss value between the output result and the fault label through the loss function;

[0028] If the loss value exceeds a preset loss threshold, the initial fault detection model is updated based on the loss value, and the loss value is recalculated.

[0029] According to the control system of the power plant air cooling device provided by the present invention, the fault detection result is also used to predict the time when the power plant air cooling device fails.

[0030] The control system of the power station air cooling device provided by the present invention further includes an alarm device;

[0031] After determining that the air cooling device of the power plant has failed, the server triggers the alarm device to sound an alarm.

[0032] The present invention also provides a control method for a power station air cooling device, comprising:

[0033] Obtain different types of performance parameters of power plant air cooling devices;

[0034] Determine the cooling efficiency of the power plant air cooling device based on the performance parameters, and optimize the power plant air cooling device when the cooling efficiency is lower than a set threshold;

[0035] Methods for optimizing power plant air cooling devices include:

[0036] The operating parameters of the power plant air cooling device are controlled based on the pre-trained optimization control model and the preset target values ​​of the performance parameters. The training set of the optimization control model is the first training set. The first training set includes the operating parameters and performance parameters of the power plant air cooling device under different working conditions. The operating parameters are the parameters used by the power plant air cooling device to control the cooling effect.

[0037] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a control method for any of the above-mentioned power station air cooling devices is implemented.

[0038] The present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the control method of any of the above-mentioned power station air cooling devices is implemented.

[0039] The present invention also provides a computer program product, comprising a computer program, and when the computer program is executed by a processor, the control method of any of the above-mentioned power station air cooling devices is implemented.

[0040] In the control system of the power plant air-cooling device provided by the present invention, the performance parameters of the power plant air-cooling device can be collected in real time through the set sensors. Since the performance parameters characterize the cooling efficiency of the power plant air-cooling device, when the cooling efficiency is determined by the server to be lower than the set threshold, the power plant air-cooling device can be optimized. Since the sensor data is collected in real time, the detection efficiency can be improved accurately and timely. In addition, the optimization process is also based on the pre-trained optimization control model for optimization. By presetting a target value for the performance parameter, the operating parameters required to achieve the target value of the performance parameter can be directly determined based on the optimization control model. In this way, the accuracy of the optimization can be further improved, so that the cooling efficiency of the power plant air-cooling device is further improved, and the overall optimal operating effect is achieved. The training set for training the optimization control model is also composed of operating parameters and performance parameters under different working conditions, which can also enable the trained power plant air-cooling device to adapt to different working conditions, and accurate air-cooling device optimization control can be performed under different environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0042] Figure 1 This is one of the structural schematic diagrams of the control system of the power plant air cooling device provided by the embodiment of the present invention;

[0043] Figure 2 This is the second structural schematic diagram of the control system of the power plant air cooling device provided by the embodiment of the present invention;

[0044] Figure 3 This is one of the flow charts of the control method of the power plant air cooling device provided by the embodiment of the present invention;

[0045] Figure 4 This is a second flow chart of a control method for a power plant air cooling device provided by an embodiment of the present invention;

[0046] Figure 5 This is a third flow chart of a control method for a power plant air cooling device provided by an embodiment of the present invention;

[0047] Figure 6 It is a schematic diagram of the physical structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] Figure 1 This is one of the structural schematic diagrams of the control system of the power plant air cooling device provided by the embodiment of the present invention.

[0050] Figure 2 This is the second structural schematic diagram of the control system of the power plant air cooling device provided by the embodiment of the present invention.

[0051] like Figure 1 and Figure 2 As shown, this embodiment provides a control system for a power plant air cooling device, including:

[0052] Several sensors are used to collect different types of performance parameters of the power plant air cooling device;

[0053] A server, configured to determine a cooling efficiency of an air cooling device of a power plant based on performance parameters, and to optimize the air cooling device of the power plant when the cooling efficiency is lower than a set threshold;

[0054] Methods for optimizing power plant air cooling devices include:

[0055] The operating parameters of the power plant air cooling device are controlled based on the pre-trained optimization control model and the preset target values ​​of the performance parameters. The training set of the optimization control model is the first training set. The first training set includes the operating parameters and performance parameters of the power plant air cooling device under different working conditions. The operating parameters are the parameters used by the power plant air cooling device to control the cooling effect.

[0056] In practical applications, the power station in this application may refer to a power station, which is a factory that converts various other forms of energy into electrical energy. The air cooling device is a device that uses ambient air as a cooling medium to cool or condense the high-temperature process fluid in the pipe. Furthermore, the power station air cooling device may refer to a heat exchanger in a power station that uses air to cool hot fluid.

[0057] The sensors in this embodiment can deploy a variety of high-precision sensors according to actual conditions, including but not limited to temperature sensors, wind speed sensors, pressure sensors, humidity sensors, etc., to ensure that the performance parameters and environmental parameters of the air cooling system can be collected in real time and accurately, where the performance parameters include but are not limited to air temperature, tube bundle temperature, wind speed, wind direction, humidity, pressure, etc.

[0058] In actual applications, servers may include data acquisition servers, intelligent control optimization servers, equipment predictive maintenance servers, data storage and processing servers, and human-computer interaction servers, which are interconnected through a high-speed and stable communication network.

[0059] The data acquisition server is used to control the sensor to collect the performance parameters of the air cooling device in real time. At the same time, the data acquisition server is also responsible for data interaction and communication with other control systems in the power station to obtain more comprehensive system operation information, so as to improve the control effect of the air cooling device;

[0060] The intelligent control optimization server can deeply analyze and process the data collected by the sensors based on the built-in algorithms and optimization models. By real-time monitoring of the system operation status and environmental changes, it can intelligently generate the best control strategy and dynamically adjust the operating parameters of the air cooling equipment, including the fan speed and shutter opening, etc. This can achieve efficient operation of the air cooling system, effectively reduce energy consumption, and significantly improve cooling efficiency.

[0061] The equipment predictive maintenance server can use machine learning technology and big data analysis methods to deeply mine equipment operation data. Through the built-in fault prediction model, it can monitor the equipment operation status and parameter change trends in real time, and accurately predict the possible failures and performance degradation of the power plant air cooling device in advance. It can send out early warning signals before the failure occurs, provide scientific basis and accurate decision support for equipment maintenance, realize preventive maintenance of equipment, effectively reduce equipment failure rate, and significantly extend equipment service life;

[0062] The data storage and processing server uses a high-performance, high-reliability database system to safely and efficiently store and manage the massive amounts of data collected, the generated control strategies, and the equipment predictive maintenance results. At the same time, it can perform pre-processing operations such as cleaning, screening, and integration on the raw data of the performance parameters collected by the sensors, providing high-quality data support for other servers in the system, ensuring the stable operation of the system and accurate decision-making;

[0063] The human-computer interaction server is equipped with an intuitive and friendly graphical user interface, which provides operators with a clear and visual display of key information such as the system's real-time operating status, control parameters, and prediction results. Operators can use this interface to easily set and adjust system operating parameters, issue control instructions, query historical data and analyze reports, achieve efficient operation and management of the system, and improve work efficiency and operating experience.

[0064] In implementation, the set threshold in this embodiment can be set manually and can be flexibly set according to specific needs in actual application. For example, when the power generation pressure of the power station is large, the air cooling device needs to have a higher cooling efficiency. At this time, the set threshold can be set to a higher value.

[0065] In the control system of the power plant air-cooling device provided in the present embodiment, the performance parameters of the power plant air-cooling device can be collected in real time through the set sensors. Since the performance parameters characterize the cooling efficiency of the power plant air-cooling device, when the cooling efficiency is determined by the server to be lower than the set threshold, the power plant air-cooling device can be optimized. Since the sensor data is collected in real time, the detection efficiency can be improved accurately and timely. In addition, the optimization process is also based on the pre-trained optimization control model for optimization. By presetting a target value for the performance parameter, the operating parameters required to achieve the target value of the performance parameter can be directly determined based on the optimization control model. In this way, the accuracy of the optimization can be further improved, so that the cooling efficiency of the power plant air-cooling device is further improved, and the overall optimal operating effect is achieved. The training set for training the optimization control model is also composed of operating parameters and performance parameters under different working conditions, which can also enable the trained power plant air-cooling device to adapt to different working conditions, and accurate air-cooling device optimization control can be performed under different environments.

[0066] In an exemplary embodiment, the optimization control model is trained by the following method:

[0067] Fuzzify the data in the first training set;

[0068] After fuzzy reasoning is performed on the performance parameters and the operating parameters, the data in the first training set is input into the optimization control initial model, and the parameters of the optimization control initial model are adjusted to determine the operating parameters corresponding to the performance parameters in the first training set;

[0069] After iteratively optimizing the performance parameters and operating parameters in the initial optimization control model, the optimization control model is output.

[0070] Fuzzy processing refers to the process of converting certain values, such as ambient temperature and unit load, into corresponding fuzzy language variable values. For example, in this embodiment, the specific temperature values ​​can be divided into fuzzy sets such as "low temperature", "medium temperature" and "high temperature". In the field of automatic control, fuzzy processing is an important link in realizing fuzzy reasoning and fuzzy control. Common fuzzification methods include graded fuzzy set method, input point membership 1 method, single point fuzzy set method and membership value method. In this embodiment, there is no restriction on the specific method of fuzzy processing. In addition, fuzzy reasoning is a pre-set fuzzy rule (a series of logical rules used by experts in the field of air cooling systems to determine the preliminary control output (operating parameters) according to the input fuzzy performance parameters in the fuzzy reasoning process based on the operating experience and professional knowledge of power plant air cooling devices). The process of reasoning based on the input variables after the above fuzzy processing to obtain the preliminary control output.

[0071] In this embodiment, by fuzzifying the data, the complexity of the data calculation process can be reduced, the calculation speed can be improved, and the purpose of adjusting the cooling effect more timely can be achieved.

[0072] In an exemplary embodiment, the iterative optimization process of optimizing the control model includes:

[0073] Compare the operating parameters output by the initial optimization control model with the pre-determined optimal operating parameters to determine the error value. The optimal operating parameters enable the power plant air cooling device to achieve the optimal cooling efficiency.

[0074] If the error value exceeds a preset error threshold, the fuzzy processing process is reversely adjusted based on the error value, and the error value is re-determined.

[0075] In practical applications, the optimization control initial model is constructed based on the existing neural network model, and according to the optimization control needs of the power plant air cooling device, the model structure, training process, input data processing, etc. are improved. The optimization control initial model can be trained based on the first training set. The optimization control initial model can output corresponding operating parameters based on the input performance parameters, that is, input a desired cooling efficiency, and the model can output how to control the power plant air cooling device. During the training process, the output operating parameters can also be compared with the preset optimal operating parameters, so that the optimization control initial model can be continuously adjusted in this way, so that the output operating parameters can be closer to the optimal operating parameters. It should be noted that the optimal operating parameters here are not fixed, but can be flexibly changed according to the different input performance parameters. In this embodiment, the optimization control initial model can be continuously adjusted and trained by presetting an optimal operating parameter, thereby improving the accuracy of the trained optimization control model.

[0076] Figure 3 This is one of the flow charts of the control method of the power plant air cooling device provided in the embodiment of the present invention.

[0077] Figure 3 The method for intelligent optimization of air cooling device based on control system is illustrated, including the following steps:

[0078] (1) Input parameters: including real-time data such as ambient temperature, wind speed, wind direction, humidity, tube bundle temperature, pressure, unit load, fan speed, shutter opening, as well as equipment performance parameters and historical operation data of the air cooling system.

[0079] (2) Data fuzzification: Convert the input precise numerical data, such as ambient temperature, unit load, etc., into fuzzy linguistic variables. For example, the temperature can be divided into fuzzy sets such as "low temperature", "medium temperature" and "high temperature".

[0080] (3) Fuzzy reasoning: Based on pre-set fuzzy rules, reasoning is performed according to the fuzzified input variables to obtain preliminary control outputs.

[0081] (4) Neural network optimization: Use neural networks to further learn and optimize the results of fuzzy reasoning and adjust the weights of the neural network to obtain more precise control output.

[0082] (5) Adaptive adjustment: Real-time monitoring of system operating status and environmental changes, and dynamic adjustment of fuzzy logic parameters (such as the shape and position of membership functions) and the learning rate of the neural network to adapt to the dynamic characteristics of the system.

[0083] (6) Output optimized control parameters such as fan speed and shutter opening to achieve efficient operation and energy-saving control of the air cooling system.

[0084] The training set consists of a large amount of air cooling system operation data under different working conditions and environmental conditions, including input parameters and corresponding optimal control parameters. The validation set is selected from data collected in different time periods and operating conditions from the training set to verify the performance and generalization ability of the algorithm on new data.

[0085] The training process of the above neural network model includes:

[0086] (1) Initialization: Randomly initialize the parameters of the fuzzy logic and the weights of the neural network.

[0087] (2) Forward calculation: The input data in the training set is processed and the control output is obtained through fuzzy reasoning and neural network calculation.

[0088] (3) Error calculation: Compare the calculated control output with the optimal control parameters in the training set and calculate the error.

[0089] (3) Reverse adjustment: adjust the parameters of fuzzy logic and weights of neural network according to the error through the back propagation algorithm.

[0090] (4) Iterative optimization: Repeat the above steps for multiple iterations until the performance of the algorithm on the validation set reaches the preset indicator or no longer improves.

[0091] The above intelligent control method has the following beneficial effects:

[0092] (1) Significantly improve cooling efficiency: Through real-time and precise control strategies, the air cooling system can always maintain the best cooling effect under various complex working conditions and environmental conditions, improve the outlet temperature stability of the cooling medium, and thus improve the power generation efficiency of the entire power station.

[0093] (2) Significantly reduce energy consumption: Dynamically adjust control parameters according to the real-time needs of the system to avoid unnecessary energy consumption. According to actual tests, it is expected to reduce system energy consumption compared with traditional control methods.

[0094] (3) Optimizing system coordination performance: Taking full account of the relationships among the components within the system, the coordinated optimization operation of components such as fans, heat exchangers, and shutters is achieved, thereby improving the overall performance and reliability of the system.

[0095] (4) Enhance system adaptability: It can quickly adapt to frequent changes in power station load, drastic fluctuations in environmental conditions, and aging and wear of equipment to ensure stable operation of the system.

[0096] In an exemplary embodiment, the server is further configured to:

[0097] Extract features of performance parameters collected by sensors based on the pre-trained fault prediction model;

[0098] The performance parameters after feature extraction are subjected to feature fusion and feature abstraction to obtain fusion features, which reflect the operating fault conditions of the air cooling device of the power station;

[0099] Based on the fused features, a fault detection result of the power plant air cooling device is output, and the fault detection result at least characterizes the fault type of the power plant air cooling device.

[0100] The traditional maintenance method for power plant air cooling devices mainly relies on regular manual inspections and planned maintenance based on time intervals. Equipment status monitoring usually only uses a limited number of monitoring methods, such as temperature monitoring and vibration monitoring, and the analysis method of monitoring data is relatively simple. However, due to the lack of effective fault prediction models and algorithms, it is impossible to accurately predict the time and type of equipment failure, resulting in the inability to detect potential early failure hazards of equipment in advance. Equipment failures often occur suddenly during operation, seriously affecting the normal production of power plants. In addition, there is a lack of comprehensive consideration of the entire life cycle of the equipment, and it is impossible to effectively predict the remaining service life and performance degradation trend of the equipment.

[0101] Based on this, the solution of this embodiment provides a method for pre-fault prediction of power plant air cooling devices. Specifically, the performance parameters collected by the sensor can be subjected to feature extraction, feature fusion, feature abstraction and other processing, so that the obtained fusion features can more accurately reflect whether there are abnormalities in the operating parameters corresponding to the performance parameters. If there is an abnormality, it means that the power plant air cooling device may have a fault, and the fault detection result is output, and the fault type can be displayed in the detection result.

[0102] In an exemplary embodiment, the training process of the fault prediction model includes:

[0103] Based on the convolutional neural network model and the recurrent neural network model, an initial fault detection model is constructed;

[0104] After initializing the parameters of the initial fault detection model, the data in the second training set is input into the initial fault detection model for model training. The data in the second training set includes performance parameters of the power plant air cooling device under different fault conditions and corresponding fault labels;

[0105] Obtain the output result of the initial fault detection model, and calculate the loss value between the output result and the fault label through the loss function;

[0106] If the loss value exceeds a preset loss threshold, the initial fault detection model is updated based on the loss value, and the loss value is recalculated.

[0107] In practical applications, convolutional neural networks (CNNs) and recurrent neural networks (RNNs) extract local features and time series features from data. CNNs are used to capture spatial correlations in data, such as specific patterns in vibration spectra, and RNNs are used to process long-term dependencies in time series data, such as the changing trend of equipment performance over time.

[0108] The loss function in this embodiment may include functions such as mean square error and cross entropy.

[0109] In practice, the second training set can be composed of a large number of historical monitoring data and corresponding actual health status labels of equipment under different operating conditions, different service life, and different maintenance levels. These data cover various states of equipment such as normal operation, minor faults, moderate faults, and serious faults, and each sample contains rich multi-source data and detailed fault information.

[0110] The output fault detection results can clearly indicate the possible fault types, such as "fan bearing wear" and "heat exchanger blockage". In practical applications, the fault detection results can also give the current health of the equipment in numerical form, such as 80% means the equipment is in good condition, and 20% means the equipment is close to failure.

[0111] In this embodiment, through advanced prediction models and analysis algorithms, it is possible to accurately predict possible equipment failures weeks or even months in advance, provide sufficient preparation time for planned maintenance, effectively avoid downtime losses caused by sudden failures, and improve equipment reliability, reduce the sudden failure rate of equipment, increase equipment availability, and ensure long-term stable operation of the power station; optimize maintenance strategies, based on accurate equipment health assessment and fault prediction, realize the transition from regular maintenance to condition-based predictive maintenance, reduce unnecessary preventive maintenance, reduce maintenance costs and equipment downtime, and save maintenance costs; extend equipment life, through effective management and maintenance of the entire life cycle of the equipment, it is possible to reasonably arrange equipment operation and maintenance plans and extend the service life of the equipment.

[0112] In an exemplary embodiment, the fault detection result is also used to predict the time when a fault occurs in the air cooling device of the power plant.

[0113] Specifically, a time window in which a failure is expected to occur can be provided, such as "a failure may occur within the next 15 days", so that the staff can be alerted in advance so that they can perform equipment maintenance in advance.

[0114] Figure 4 This is the second flow chart of the control method of the power plant air cooling device provided in the embodiment of the present invention.

[0115] Figure 4The method of predicting equipment failure based on control system is illustrated. Figure 4 As shown, the following steps are included:

[0116] (1) Input parameters: including the equipment’s real-time operating time, speed, temperature, pressure, vibration spectrum, current, voltage, sound characteristics, oil chemical indicators and other real-time monitoring parameters, as well as the equipment’s historical maintenance records, parts replacement records, equipment model, installation date and other related information.

[0117] (2) Data preprocessing: Clean, denoise, standardize and normalize the input multi-source heterogeneous data to ensure data quality and consistency.

[0118] (3) Feature extraction: Local features and time series features in the data are extracted through multi-layer convolutional neural networks (CNNs) and recurrent neural networks (RNNs). CNNs are used to capture spatial correlations in data, such as specific patterns in vibration spectra; RNNs are used to process long-term dependencies in time series data, such as the changing trend of equipment performance over time.

[0119] (4) Feature fusion and abstraction: The extracted features are fused and abstracted, and the features are combined and compressed through the fully connected layer to obtain a comprehensive feature representation that can reflect the health status of the device.

[0120] (5) Prediction output: Based on the fused features, the output layer performs the final prediction, outputting the health status assessment value of the equipment in the future (expressed as a percentage), possible fault types (such as mechanical wear, electrical failure, poor lubrication, etc.), and the expected time range of the fault occurrence (accurate to days or hours).

[0121] Among them, the health status assessment can give the current health status of the equipment in numerical form, such as 80% means that the equipment is in good condition, and 20% means that the equipment is close to a fault state; the fault type diagnosis can clearly point out the possible types of faults, such as "fan bearing wear" and "heat exchanger blockage". The fault occurrence time prediction can provide a time window in which the fault is expected to occur, such as "a fault may occur within the next 15 days."

[0122] In addition, the training set can be composed of historical monitoring data of a large number of equipment under different operating conditions, different service life, and different maintenance levels, and the corresponding actual health status labels. These data cover various states such as normal operation, minor faults, moderate faults, and serious faults of the equipment, and each sample contains rich multi-source data and detailed fault information. The validation set is selected from the data collected in different time periods and under different operating conditions from the training set to verify the performance and generalization ability of the model on unseen data. The data in the validation set also has accurate health status labels and fault information.

[0123] The following training process is performed based on the training set and the validation set:

[0124] (1) Initialize model parameters: Randomly initialize the weights, biases and other parameters of CNN and RNN.

[0125] (2) Forward propagation: The sample data in the training set is input into the initial fault detection model constructed by the convolutional neural network (CNN) and the recurrent neural network (RNN) to calculate the prediction results.

[0126] (3) Calculate loss: Calculate the error between the predicted result and the actual label through the loss function (such as mean square error, cross entropy, etc.).

[0127] (4) Back propagation: Based on the gradient of the loss function, the model parameters such as weights and biases are updated through the back propagation algorithm.

[0128] (5) Iterative optimization: Repeat the above steps for multiple iterations until the performance of the model on the validation set reaches the preset indicators (such as accuracy reaching more than 90%, recall reaching more than 80%, etc.) or no longer improves.

[0129] The above control method has the following beneficial effects:

[0130] (1) Predicting failures in advance: Advanced prediction models and analysis algorithms can accurately predict possible equipment failures weeks or even months in advance, providing sufficient preparation time for planned maintenance and effectively avoiding downtime losses caused by sudden failures.

[0131] (2) Improve equipment reliability: Reduce the sudden failure rate of equipment, increase equipment availability, and ensure long-term stable operation of the power station.

[0132] (3) Optimizing maintenance strategies: Based on accurate equipment health assessment and fault prediction, the system achieves a shift from scheduled maintenance to condition-based predictive maintenance, reduces unnecessary preventive maintenance, reduces maintenance costs and equipment downtime, and saves maintenance costs.

[0133] (4) Extending equipment life: Through effective management and maintenance of the equipment throughout its life cycle, the operation and maintenance plans of the equipment can be reasonably arranged to extend the service life of the equipment.

[0134] In an exemplary embodiment, an alarm device is also included;

[0135] After determining that the air cooling device of the power plant has failed, the server triggers the alarm device to sound an alarm.

[0136] In practical applications, we can combine expert knowledge and rich historical fault data to set reasonable warning thresholds and fault diagnosis rules. When the equipment health indicators predicted by the model are close to or exceed the warning threshold, the system can automatically issue a warning signal and provide a detailed fault diagnosis report, including possible fault types, fault locations, fault severity and recommended maintenance measures.

[0137] The control method of the power plant air cooling device provided by the present invention is described below. The control method of the power plant air cooling device described below and the control system of the power plant air cooling device described above can be referred to each other.

[0138] Figure 5 This is the third flow chart of the control method of the power plant air cooling device provided in the embodiment of the present invention.

[0139] like Figure 5 As shown, the control method provided in this embodiment includes:

[0140] Step 501, obtaining different types of performance parameters of the power station air cooling device;

[0141] Step 502, determining the cooling efficiency of the power plant air cooling device based on the performance parameter, and optimizing the power plant air cooling device when the cooling efficiency is lower than a set threshold;

[0142] Methods for optimizing power plant air cooling devices include:

[0143] The operating parameters of the power plant air cooling device are controlled based on the pre-trained optimization control model and the preset target values ​​of the performance parameters. The training set of the optimization control model is the first training set. The first training set includes the operating parameters and performance parameters of the power plant air cooling device under different working conditions. The operating parameters are the parameters used by the power plant air cooling device to control the cooling effect.

[0144] In an exemplary embodiment, the optimization control model is trained by the following method:

[0145] Fuzzify the data in the first training set;

[0146] After fuzzy reasoning is performed on the performance parameters and the operating parameters, the data in the first training set is input into the optimization control initial model, and the parameters of the optimization control initial model are adjusted to determine the operating parameters corresponding to the performance parameters in the first training set;

[0147] After iteratively optimizing the performance parameters and operating parameters in the initial optimization control model, the optimization control model is output.

[0148] In an exemplary embodiment, the iterative optimization process of optimizing the control model includes:

[0149] Performing fuzzy processing on the data in the first training set, where the first training set includes performance parameters and operating parameters of the power plant air cooling device under different operating conditions;

[0150] Inputting the data in the first training set into the optimization control initial model to determine the operating parameters corresponding to the performance parameters in the first training set;

[0151] Compare the operating parameters output by the initial optimization control model with the pre-determined optimal operating parameters to determine the error value. The optimal operating parameters enable the power plant air cooling device to achieve the optimal cooling efficiency.

[0152] If the error value exceeds a preset error threshold, the fuzzy processing process is reversely adjusted based on the error value, and the error value is re-determined.

[0153] In an exemplary embodiment, the server is further configured to:

[0154] Extract features of performance parameters collected by sensors based on the pre-trained fault prediction model;

[0155] The performance parameters after feature extraction are subjected to feature fusion and feature abstraction to obtain fusion features, which reflect the operating fault conditions of the air cooling device of the power station;

[0156] Based on the fused features, a fault detection result of the power plant air cooling device is output, and the fault detection result at least characterizes the fault type of the power plant air cooling device.

[0157] In an exemplary embodiment, the training process of the fault prediction model includes:

[0158] Based on the convolutional neural network model and the recurrent neural network model, an initial fault detection model is constructed;

[0159] After initializing the parameters of the initial fault detection model, the data in the second training set is input into the initial fault detection model for model training. The data in the second training set includes performance parameters of the power plant air cooling device under different fault conditions and corresponding fault labels;

[0160] Obtain the output result of the initial fault detection model, and calculate the loss value between the output result and the fault label through the loss function;

[0161] If the loss value exceeds a preset loss threshold, the initial fault detection model is updated based on the loss value, and the loss value is recalculated.

[0162] In an exemplary embodiment, the fault detection result is also used to predict the time when a fault occurs in the air cooling device of the power plant.

[0163] In an exemplary embodiment, after determining that the air cooling device of the power plant has failed, the server triggers the alarm device to sound an alarm.

[0164] The specific implementation method of the control method of the power plant air cooling device provided in this embodiment can be implemented with reference to the above embodiment, and will not be described in detail here.

[0165] The control system and method of the above-mentioned power plant air cooling device provided in this application can (1) improve the operating efficiency and reliability of the power plant: through intelligent control optimization and equipment predictive maintenance, the cooling efficiency of the power plant air cooling system can be significantly improved, energy consumption can be reduced, the occurrence rate of equipment failures can be reduced, and the service life of the equipment can be extended, thereby improving the overall operating efficiency and reliability of the power plant. This is of great significance for ensuring the stability and safety of power supply. (2) Adapt to the needs of energy transformation: With the rapid development of renewable energy, the power system has increasingly higher requirements for flexibility and regulation capabilities. This patent can help the power plant air cooling system better adapt to the frequent changes in unit loads, improve the system's response speed and regulation capabilities, and provide strong support for energy transformation. (3) Reduce operating costs: Accurate equipment predictive maintenance can reduce unnecessary preventive maintenance, avoid excessive or insufficient maintenance, thereby reducing maintenance costs and equipment downtime, and improving the economic benefits of the power plant. At the same time, intelligent control optimization can reduce system energy consumption and further save operating costs. (4) Promote application in other fields: The intelligent control algorithm, machine learning technology, and big data analysis method used in this patent have certain versatility and can be promoted and applied to other similar industrial cooling systems or related fields, providing reference and reference for the intelligent development of these fields. (5) Meet environmental protection requirements: Efficient air cooling system operation can reduce energy consumption and greenhouse gas emissions, which is in line with current environmental protection requirements and the trend of sustainable development.

[0166] In summary, the solution of this application has good application prospects and will bring significant benefits and driving force to the intelligent development of power plant air cooling systems.

[0167] Figure 6 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 6 As shown, the electronic device may include: a processor (processor) 610, a communication interface (Communications Interface) 620, a memory (memory) 630 and a communication bus 640, wherein the processor 610, the communication interface 620, and the memory 630 communicate with each other through the communication bus 640. The processor 610 may call the logic instructions in the memory 630 to execute the control method of the power plant air cooling device, and the method includes:

[0168] Obtaining performance parameters of the power plant air cooling device, which characterize the cooling efficiency of the power plant air cooling device;

[0169] When the performance parameters are lower than the set threshold, the power plant air cooling device is optimized;

[0170] Methods for optimizing power plant air cooling devices include:

[0171] The operating parameters of the power plant air cooling device are controlled based on a predetermined first corresponding relationship and a preset target value of a performance parameter. The first corresponding relationship is the corresponding relationship between the performance parameter and the operating parameter of the power plant air cooling device. The operating parameter is a parameter of the power plant air cooling device used to control the cooling effect.

[0172] In addition, the logic instructions in the above-mentioned memory 630 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods of various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0173] On the other hand, the present invention further provides a computer program product, the computer program product includes a computer program, the computer program can be stored in a non-transitory computer-readable storage medium, when the computer program is executed by a processor, the computer can execute the control method of the power plant air cooling device provided by the above methods, the method includes:

[0174] Obtaining performance parameters of the power plant air cooling device, which characterize the cooling efficiency of the power plant air cooling device;

[0175] When the performance parameters are lower than the set threshold, the power plant air cooling device is optimized;

[0176] Methods for optimizing power plant air cooling devices include:

[0177] The operating parameters of the power plant air cooling device are controlled based on a predetermined first corresponding relationship and a preset target value of a performance parameter. The first corresponding relationship is the corresponding relationship between the performance parameter and the operating parameter of the power plant air cooling device. The operating parameter is a parameter of the power plant air cooling device used to control the cooling effect.

[0178] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the control method of the power plant air cooling device provided by the above methods is implemented, and the method includes:

[0179] Obtaining performance parameters of the power plant air cooling device, which characterize the cooling efficiency of the power plant air cooling device;

[0180] When the performance parameters are lower than the set threshold, the power plant air cooling device is optimized;

[0181] Methods for optimizing power plant air cooling devices include:

[0182] The operating parameters of the power plant air cooling device are controlled based on a predetermined first corresponding relationship and a preset target value of a performance parameter. The first corresponding relationship is the corresponding relationship between the performance parameter and the operating parameter of the power plant air cooling device. The operating parameter is a parameter of the power plant air cooling device used to control the cooling effect.

[0183] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, i.e., they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Those of ordinary skill in the art may understand and implement it without creative effort.

[0184] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment or some parts of the embodiment.

[0185] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. Control system of power station air cooling device, It is characterized in that include: Several sensors are used to collect different types of performance parameters of the power plant air cooling device; A server, configured to determine a cooling efficiency of the power plant air cooling device based on the performance parameter, and to optimize the power plant air cooling device when the cooling efficiency is lower than a set threshold; The method for optimizing the power station air cooling device includes: The operating parameters of the power plant air cooling device are controlled based on the pre-trained optimization control model and the preset target values ​​of the performance parameters. The training set of the optimization control model is a first training set, which includes the operating parameters and performance parameters of the power plant air cooling device under different working conditions. The operating parameters are the parameters used by the power plant air cooling device to control the cooling effect.

2. The control system of the power plant air cooling device according to claim 1, It is characterized in that The optimization control model is trained by the following method: Performing fuzzy processing on the data in the first training set; After fuzzy reasoning is performed on the performance parameters and the operating parameters, the data in the first training set is input into an initial optimization control model, and the parameters of the initial optimization control model are adjusted to determine the operating parameters corresponding to the performance parameters in the first training set; After iteratively optimizing the performance parameters and the operating parameters in the optimization control initial model, the optimization control model is outputted.

3. The control system of the power plant air cooling device according to claim 2, It is characterized in that The iterative optimization process of the optimization control model includes: Comparing the operating parameters output by the optimization control initial model with the pre-determined optimal operating parameters to determine an error value, wherein the optimal operating parameters enable the power station air cooling device to achieve an optimal cooling efficiency; If the error value exceeds a preset error threshold, the fuzzy processing process is reversely adjusted based on the error value, and the error value is re-determined.

4. The control system of the power plant air cooling device according to claim 1, It is characterized in that The server is also used to: Extracting features of the performance parameters collected by the sensor based on a pre-trained fault prediction model; Performing feature fusion and feature abstraction on the performance parameters after feature extraction to obtain fusion features, wherein the fusion features reflect the operating fault conditions of the air cooling device of the power station; Based on the fusion feature, a fault detection result of the power plant air cooling device is output, and the fault detection result at least represents the fault type of the power plant air cooling device.

5. The control system of the power plant air cooling device according to claim 4, It is characterized in that The training process of the fault prediction model includes: Based on the convolutional neural network model and the recurrent neural network model, an initial fault detection model is constructed; After initializing the parameters of the initial fault detection model, inputting the data in the second training set into the initial fault detection model for model training, wherein the data in the second training set includes performance parameters of the air cooling device of the power station under different fault conditions and corresponding fault labels; Obtaining an output result of the initial fault detection model, and calculating a loss value between the output result and the fault label using a loss function; If the loss value exceeds a preset loss threshold, the fault detection initial model is updated based on the loss value, and the loss value is recalculated.

6. The control system of the power plant air cooling device according to claim 4, It is characterized in that The fault detection result is also used to predict the time when the power station air cooling device fails.

7. The control system of the power plant air cooling device according to claim 4, It is characterized in that It also includes an alarm device; After determining that the air cooling device of the power station has failed, the server triggers the alarm device to sound an alarm.

8. Control method of power station air cooling device, It is characterized in that include: Obtain different types of performance parameters of power plant air cooling devices; determining a cooling efficiency of the power plant air cooling device based on the performance parameter, and optimizing the power plant air cooling device when the cooling efficiency is lower than a set threshold; The method for optimizing the power station air cooling device includes: The operating parameters of the power plant air cooling device are controlled based on the pre-trained optimization control model and the preset target values ​​of the performance parameters. The training set of the optimization control model is a first training set, which includes the operating parameters and performance parameters of the power plant air cooling device under different working conditions. The operating parameters are the parameters used by the power plant air cooling device to control the cooling effect.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the program, the control method of the power plant air cooling device as claimed in claim 8 is implemented.

10. A computer-readable storage medium storing a computer program, It is characterized in that When the computer program is executed by the processor, the control method of the power plant air cooling device as claimed in claim 8 is implemented.

Images