Condenser vacuum degree value prediction method and device, processor and storage medium
By selecting parameters with high correlation to vacuum level, and using an LSTM model to train a method for predicting condenser vacuum level, the problems of long training time and low accuracy in existing technologies are solved, achieving efficient and accurate vacuum level prediction and ensuring the optimized operation of the cold end system.
Patent Information
- Application Number
- CN202411120996.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2024-08-02
- Filing Date
- 2024-08-15
- Publication Date
- 2025-01-14
AI Technical Summary
Existing methods for predicting condenser vacuum levels suffer from long model training times and low prediction accuracy due to the large number of operating parameters in the cold-end system.
By selecting target-related parameters that are highly correlated with vacuum level, and using the LSTM model for model training, the influence of irrelevant parameters is reduced, thereby improving training efficiency and accuracy.
This improves the accuracy and reliability of condenser vacuum prediction, ensuring stable operation of the cold end system.
Smart Images
Figure CN119312084A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of condensers, and in particular to a method for predicting a condenser vacuum value, a device for predicting a condenser vacuum value, a processor, and a storage medium. Background Art
[0002] The cold end system in a thermal power generating unit is a thermal system composed of condensers, circulating water pumps and other equipment. When a problem occurs in the cold end system, the output of the unit will be affected through the condenser vacuum, reducing the unit's circulation efficiency, and ultimately leading to an increase in the power supply coal consumption rate, affecting the unit's thermal economy. Based on this, ensuring that the condenser operates at or near the optimal vacuum value and achieving optimal operation of the cold end system is an important means for power plants to save energy and reduce consumption and improve the thermal economy of the unit.
[0003] In the related art, in order to predict the condenser vacuum value, the historical parameter values of each operating parameter when the cold end system is running are usually used to train the model, and then the values of each operating parameter are obtained, and the condenser vacuum value is predicted using the trained model. Although this method can predict the condenser vacuum value, due to the large number of operating parameters involved in the operation of the cold end system, the number of values contained in a single training sample during model training is large, and there are values of operating parameters that have little or no effect on the condenser vacuum value, resulting in a long model training time and low prediction accuracy of the trained model. Summary of the invention
[0004] The purpose of the embodiments of the present application is to provide a condenser vacuum value prediction method, a condenser vacuum value prediction device, a processor and a storage medium to solve the problem of low prediction accuracy of the existing condenser vacuum value prediction method.
[0005] In order to achieve the above-mentioned object, the first aspect of the present application provides a method for predicting a condenser vacuum value, the method comprising: Obtaining a target data sequence for each target-related parameter in a target-related parameter set for the condenser; wherein the target-related parameter is an operating parameter whose correlation value with the vacuum degree in the cold end system satisfies a preset correlation condition; Each target data sequence is input into a preset target condenser vacuum value prediction model to obtain a target prediction result of the condenser vacuum.
[0006] In the embodiment of the present application, before inputting each target data sequence into the preset target condenser vacuum value prediction model, the method further includes: Acquire a historical data set; wherein the historical data set includes a plurality of historical data subsets; each of the historical data subsets includes a historical value of each operating parameter corresponding to a plurality of historical moments, and a historical value of a condenser vacuum degree at least one future moment corresponding to the plurality of historical moments; Based on the preset steady-state condition, a steady-state screening is performed on a historical data subset in the historical data set to obtain multiple steady-state data subsets; The preset initial condenser vacuum value prediction model is trained using the steady-state data subset to obtain the preset target condenser vacuum value prediction model.
[0007] In the embodiment of the present application, before the steady-state data subset is used to perform model training on the preset initial condenser vacuum value prediction model, the method further includes: For each operating parameter, based on the historical value of the operating parameter and the historical value of the condenser vacuum in each steady-state data subset, the Spearman correlation coefficient is used to determine the correlation value corresponding to the operating parameter; For each operating parameter, if the correlation value corresponding to the operating parameter meets the preset correlation condition, the operating parameter is determined to be a target correlation parameter.
[0008] In the embodiment of the present application, a preset initial condenser vacuum value prediction model is trained using a steady-state data subset to obtain a preset target condenser vacuum value prediction model, including: Determine a training sample set and a test sample set according to each steady-state data subset; wherein the training sample set includes a plurality of first steady-state data subsets, and the test sample set includes a plurality of second steady-state data subsets; For each first steady-state data subset, the historical values of each target-related parameter in the first steady-state data subset are used as model input, and the historical values of the condenser vacuum are used as model output, and the preset initial condenser vacuum value prediction model is trained to obtain a pending model; For each second steady-state data subset, the historical values of each target-related parameter in the second steady-state data subset are used as inputs of the prediction model to be evaluated to obtain a prediction result corresponding to the second steady-state data subset; According to the historical values of the condenser vacuum and the corresponding prediction results in each second steady-state data subset, determine whether to use the test sample set to train the model to be determined. If so, use the test sample set to train the model to determine the preset target condenser vacuum value prediction model. If not, determine the model to be determined as the preset target condenser vacuum value prediction model.
[0009] In the embodiment of the present application, determining whether to use the test sample set to perform model training on the model to be determined according to the historical values of the condenser vacuum degree and the corresponding prediction results in each second steady-state data subset includes: Determine the overall prediction error value of the undetermined model according to the historical value of the condenser vacuum degree in each second steady-state data subset and the prediction result corresponding to each second steady-state data subset; According to the overall prediction error value and the preset error threshold, determine whether to use the test sample set to train the model to be determined.
[0010] In the embodiment of the present application, after inputting each target data sequence into a preset target condenser vacuum value prediction model to obtain a target prediction result of the condenser vacuum, the method further includes: Based on the target prediction result of the condenser vacuum and the preset vacuum reference value range, determine whether it is necessary to enter the vacuum abnormality warning mode.
[0011] In an embodiment of the present application, the target-related parameter set includes circulating water outlet temperature, hot well condensate temperature, low-pressure cylinder exhaust temperature, unit load, condensate flow rate and circulating water inlet temperature.
[0012] A second aspect of the present application provides a condenser vacuum value prediction device, the device comprising: A relevant parameter value acquisition module is used to obtain a target data sequence for each target relevant parameter in a target relevant parameter set for the condenser; wherein the target relevant parameter is an operating parameter whose correlation value with the vacuum degree in the cold end system meets a preset correlation condition; The vacuum value prediction module is used to input each target data sequence into a preset target condenser vacuum value prediction model to obtain a target prediction result of the condenser vacuum.
[0013] A third aspect of the present application provides a processor configured to execute the above-mentioned condenser vacuum value prediction method.
[0014] A fourth aspect of the present application provides a machine-readable storage medium having instructions stored thereon, which, when executed by a processor, configure the processor to execute the above-mentioned condenser vacuum value prediction method.
[0015] The condenser vacuum value prediction method, device, processor and storage medium provided in the present application only use the historical values of the target related parameters whose correlation values with the vacuum degree in the cold end system meet the preset correlation conditions for model training, thereby reducing the influence of the values of the operating parameters that have little or no influence on the condenser vacuum value on the model's condenser vacuum value prediction, thereby improving the model training efficiency and the prediction accuracy of the trained model; subsequently, the target data sequence of multiple parameter values of the target related parameters arranged in corresponding chronological order is directly input into the trained model (i.e., the preset target condenser vacuum value prediction model) to predict the condenser vacuum value, thereby ensuring the high accuracy and high reliability of the condenser vacuum value prediction.
[0016] Other features and advantages of the embodiments of the present application will be described in detail in the subsequent specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present application, but do not constitute a limitation on the embodiments of the present application. In the accompanying drawings: Figure 1 The schematic diagram of the flow chart of the condenser vacuum value prediction method according to the embodiment of the present application is shown; Figure 2 A structural block diagram of a condenser vacuum value prediction device according to an embodiment of the present application is schematically shown; Figure 3 The internal structure diagram of the computer device according to the embodiment of the present application is schematically shown.
[0018] Description of Reference Numerals A01-processor; A02-network interface; A03-internal memory; A04-display screen; A05-input device; A06-non-volatile storage medium; B01-operating system; B02-computer program. DETAILED DESCRIPTION
[0019] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the specific implementation methods described herein are only used to illustrate and explain the embodiments of the present application, and are not used to limit the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.
[0020] It should be noted that if the embodiments of the present application involve directional indications (such as up, down, left, right, front, back, etc.), such directional indications are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.
[0021] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present application, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or suggesting their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by this application.
[0022] In view of the problem of low prediction accuracy of the condenser vacuum value prediction method in the related art, the present application provides a condenser vacuum value prediction method, device, processor and storage medium. The condenser vacuum value prediction method, device, processor and storage medium provided in the present application are described in detail below in combination with the accompanying drawings through specific examples and implementation methods.
[0023] Figure 1 The flowchart of the method for predicting the condenser vacuum value according to the embodiment of the present application is schematically shown. Figure 1 As shown, in one embodiment of the present application, a method for predicting a condenser vacuum value is provided, the method comprising: Step 200: Obtain a target data sequence for each target-related parameter in a target-related parameter set for a condenser.
[0024] Among them, the target related parameter is an operating parameter whose correlation value with the vacuum degree in the cold end system meets the preset correlation condition, and the target data sequence of the target related parameter is obtained by arranging multiple parameter values of the target related parameter in corresponding chronological order.
[0025] In this embodiment, the preset correlation condition is that the correlation value is greater than the preset correlation threshold. That is, for each operating parameter of the cold end system, if the correlation value between the operating parameter and the vacuum degree is greater than the preset correlation threshold, the operating parameter is the target correlation parameter.
[0026] In this embodiment, the target-related parameter set includes circulating water outlet temperature, hot well condensate temperature, low-pressure cylinder exhaust temperature, unit load, condensate flow rate and circulating water inlet temperature.
[0027] Step 400: Input each target data sequence into a preset target condenser vacuum value prediction model to obtain a target prediction result of the condenser vacuum.
[0028] The preset target condenser vacuum value prediction model is determined based on a training sample set and a test sample set determined under preset steady-state conditions.
[0029] In addition, the prediction result obtained by using the model for prediction may be a single value or a sequence of values. That is to say, the target prediction result may be the predicted value of the condenser vacuum at a certain moment in the future, or it may be a prediction sequence of the condenser vacuum in a certain time period in the future (which includes multiple predicted values of the condenser vacuum). In this embodiment, whether the target prediction result is a single value or a sequence of values is determined by the preset target condenser vacuum value prediction model.
[0030] In this embodiment, in order to obtain a preset target condenser vacuum value prediction model, before step 400, the method further includes: Step 120, obtaining a historical data set.
[0031] The historical data set includes multiple historical data subsets; each historical data subset includes historical values of each operating parameter corresponding to multiple historical moments, and historical values of condenser vacuum at at least one future moment corresponding to the multiple historical moments.
[0032] Specifically, the operating parameters are operating parameters involved in the operation of the cold end system, and the future moment corresponding to the historical moment represents the moment after the historical moment. This embodiment obtains multiple historical data subsets to determine a steady-state data subset for model training and to determine operating parameters with a high correlation with the condenser vacuum, i.e., target-related parameters.
[0033] It should be noted that, in this embodiment, each historical data subset contains an equal number of historical values of condenser vacuum, and the number is determined according to actual needs. When the historical data subset contains only one historical value of condenser vacuum, the target prediction result output by the preset target condenser vacuum value prediction model is the predicted value of the condenser vacuum at a certain moment in the future. When the historical data subset contains multiple historical values of condenser vacuum, the target prediction result output by the preset target condenser vacuum value prediction model is the prediction sequence of the condenser vacuum in a certain time period in the future.
[0034] Step 140 , based on a preset steady-state condition, a steady-state screening is performed on a historical data subset in the historical data set to obtain a plurality of steady-state data subsets.
[0035] The steady-state data subset is a historical data subset contained in the historical data set.
[0036] In one example, the preset steady-state condition is: for each preset target parameter in the preset target parameter set, within the preset unit time, the difference between the maximum value and the minimum value of the preset target parameter is less than the judgment threshold corresponding to the preset target parameter. Among them, the preset target parameter set includes unit load, main steam pressure, main steam temperature, reheat steam temperature and feed water flow. The preset unit time can be selected as 30min, and the judgment thresholds corresponding to the unit load, main steam pressure, main steam temperature, reheat steam temperature and feed water flow are selected as 15MW, 3MPa, 15℃, 15℃, and 80t / h respectively. It is worth mentioning that each preset target parameter in the preset target parameter set is an operating parameter involved in the operation of the cold end system.
[0037] In this example, the historical data subset in the historical data set is screened in a steady state through the window sliding detection technology. Specifically, within the preset unit time, when the difference between the maximum and minimum values of the operating parameters such as unit load, main steam pressure, main steam temperature, reheat steam temperature and feed water flow is less than the corresponding judgment threshold, the historical data subset within this time window is a steady-state data subset, the window slides backward, and the judgment continues.
[0038] Step 160 , using the steady-state data subset to perform model training on a preset initial condenser vacuum value prediction model to obtain a preset target condenser vacuum value prediction model.
[0039] In this embodiment, before step 160, the method further includes: Step 151: for each operating parameter, based on the historical value of the operating parameter and the historical value of the condenser vacuum in each steady-state data subset, the Spearman correlation coefficient is used to determine the correlation value corresponding to the operating parameter.
[0040] The correlation value corresponding to the operating parameter is the correlation value between the operating parameter and the vacuum degree, and the numerical value of the correlation value reflects the strength of the correlation between the operating parameter and the vacuum degree.
[0041] Specifically, when the spearman correlation coefficient is used to determine the correlation value between the operating parameter and the vacuum degree, the calculation result range is [-1,1], where a negative value indicates that the operating parameter and the vacuum degree are negatively correlated, and a positive value indicates that the operating parameter and the vacuum degree are positively correlated. This embodiment uses the spearman correlation coefficient to determine the correlation value corresponding to each operating parameter affecting the condenser vacuum degree obtained by the mechanism analysis, to further select the key parameters affecting the condenser vacuum degree (i.e., the target related parameters), so as to ensure that the dimension of the model input is reduced as much as possible without losing sample information, and to improve the training efficiency and accuracy of the model.
[0042] Step 152: for each operating parameter, if the correlation value corresponding to the operating parameter satisfies a preset correlation condition, the operating parameter is determined to be a target correlation parameter.
[0043] Among them, for each operating parameter, when the correlation value corresponding to the operating parameter (denoted as )satisfy When , the correlation between the operating parameter and the vacuum degree is very strong; when When , it indicates that the correlation between the operating parameter and the vacuum degree is strong; when When , it indicates that the correlation between the operating parameter and the vacuum degree is moderate; when When , the correlation between the operating parameter and the vacuum degree is weak. , indicating that the correlation between the operating parameter and the vacuum degree is extremely weak or unrelated.
[0044] In this embodiment, some operating parameters (whose values are variables) involved in the operation of the cold end system and their corresponding correlation values are shown in Table 1.
[0045] Table 1 - Operation parameters and their corresponding correlation values Operation parameters Correlation value Circulating water outlet temperature 0.990132 Hot well condensate temperature 0.980651 Low pressure cylinder exhaust temperature 0.968569 Unit load 0.864817 Condensate flow 0.860158 Circulating water inlet temperature 0.642832 Vacuum pump current -0.327932 Circulating water pump current 0.322974 Hot well water replenishment flow 0.045632 In one example, the preset correlation condition is that the correlation value is greater than 0.6. At this time, combined with Table 1, it can be concluded that the target related parameter set includes circulating water outlet temperature, hot well condensate temperature, low pressure cylinder exhaust temperature, unit load, condensate flow rate and circulating water inlet temperature.
[0046] In this embodiment, step 160 includes: Step 161: Determine a training sample set and a test sample set according to each steady-state data subset.
[0047] The training sample set includes a plurality of first steady-state data subsets, and the test sample set includes a plurality of second steady-state data subsets.
[0048] The first steady-state data subset and the second steady-state data subset are both steady-state data subsets included in the historical data set, that is, historical data subsets included in the historical data set.
[0049] In one example, the ratio of the number of the first steady-state data subset in the training sample set to the number of the second steady-state data subset in the test sample set is selected to be 7: 3. Of course, it can be understood that the ratio of the two can also be adjusted accordingly according to actual conditions, and this embodiment does not specifically limit it.
[0050] Step 162, for each first steady-state data subset, the historical values of each target-related parameter in the first steady-state data subset are used as model input, and the historical values of condenser vacuum are used as model output, and the preset initial condenser vacuum value prediction model is trained to obtain a pending model.
[0051] In this embodiment, the preset initial condenser vacuum value prediction model adopts a long short-term memory network (Long Short-Term Memory, LSTM) model. Among them, the LSTM model can solve the gradient vanishing or gradient exploding problems that may occur in the traditional recurrent neural network (Recurrent Neural Network, RNN) model when processing long sequences by introducing a gating mechanism, thereby better capturing long-term dependencies.
[0052] Step 163: for each second steady-state data subset, the historical values of each target-related parameter in the second steady-state data subset are used as inputs of the prediction model to be evaluated to obtain a prediction result corresponding to the second steady-state data subset.
[0053] Among them, for each second steady-state data subset, the prediction result corresponding to the second steady-state data subset includes at least one predicted value of the condenser vacuum, and the number of predicted values of the condenser vacuum included in the prediction result is equal to the number of historical values of the condenser vacuum in the second steady-state data subset. Simply put, if the model outputs a single value during the model training process, then the model outputs a single value when predicting. Similarly, if the model outputs a sequence of values arranged in a corresponding chronological order by multiple values during the model training process, then the model outputs a sequence of values arranged in a corresponding chronological order by multiple values when predicting.
[0054] Step 164, based on the historical values of the condenser vacuum in each second steady-state data subset and the corresponding prediction results, determine whether to use the test sample set to train the model to be determined. If so, use the test sample set to train the model to be determined to obtain a preset target condenser vacuum value prediction model. If not, determine the model to be determined as the preset target condenser vacuum value prediction model.
[0055] In a specific implementation, the step 164 specifically includes: Step 1641 , determining the overall prediction error value of the pending model according to the historical values of the condenser vacuum in each second steady-state data subset and the prediction results corresponding to each second steady-state data subset.
[0056] Specifically, when the prediction results corresponding to each second steady-state data subset only contain one prediction value, the overall prediction error value of the undetermined model is determined by the root mean square error (RMSE), mean absolute error (MAE) or mean absolute percentage error (MAPE) based on the historical values of the condenser vacuum in each second steady-state data subset and the prediction results corresponding to each second steady-state data subset.
[0057] When the prediction results corresponding to each second steady-state data subset contain multiple prediction values, for each prediction result, determine the absolute value of the error between each prediction value in the prediction result and the corresponding historical value of the condenser vacuum, and determine the prediction error value corresponding to the second steady-state data subset corresponding to the prediction result by the arithmetic mean method; based on each prediction error value, determine the overall prediction error value of the model to be determined by the arithmetic mean method.
[0058] It can be understood that, for each second steady-state data subset, each historical value of the condenser vacuum in the second steady-state data subset and each predicted value in the prediction result corresponding to the second steady-state data subset correspond based on a time point.
[0059] Step 1642, based on the overall prediction error value and the preset error threshold, determine whether to use the test sample set to perform model training on the model to be determined.
[0060] Specifically, when the overall prediction error value is greater than or equal to the preset error threshold, it can be considered that the prediction accuracy of the pending model obtained by the current training has reached the expected level and meets the actual requirements. At this time, there is no need for further model optimization, and the model training can be ended, and the pending model can be used as the preset target condenser vacuum value prediction model; when the overall prediction error value is less than the preset error threshold, it can be considered that the prediction accuracy of the pending model obtained by the current training can be further improved. At this time, the test sample set can be used to continue model training for the pending model, thereby further improving the accuracy of the model prediction.
[0061] In this embodiment, the method of using the test sample set to perform model training on the model to be determined is the same as the method of using the training sample set to perform model training on the preset initial condenser vacuum value prediction model, which will not be described in detail here.
[0062] After the preset initial condenser vacuum value prediction model is trained using the training sample set, this embodiment ensures the accuracy of model prediction by evaluating whether to further train the undetermined model obtained through training.
[0063] In this embodiment, step 400 can obtain a target prediction result of the condenser vacuum at a certain moment in the future or a certain time period in the future by inputting the acquired target data sequences into a preset target condenser vacuum value prediction model. Based on the target prediction result, it can be convenient to subsequently determine whether it is necessary to adjust the operating state of at least part of the cold end system so that the condenser operates at an optimal vacuum value or near the optimal vacuum value, thereby ensuring stable operation of the cold end system.
[0064] To determine whether it is necessary to adjust the operating state of at least part of the components of the cold end system, optionally, the method further includes: Step 600: Determine whether to enter a vacuum abnormality warning mode based on the target prediction result of the condenser vacuum and a preset vacuum reference value interval.
[0065] Specifically, when the target prediction result of the condenser vacuum degree belongs to the preset vacuum reference value interval, it indicates that the value of the condenser vacuum degree will be within the normal range at a certain moment in the future or at a certain time period in the future, that is, the condenser will be operating at or near the optimal vacuum value at a certain moment in the future or at a certain time period in the future. At this time, there is no need to enter the vacuum abnormality warning mode, that is, there is no need to adjust the operating status of the components of the cold end system.
[0066] When the target prediction result of the condenser vacuum degree does not fall within the preset vacuum reference value interval, it indicates that the condenser vacuum degree value will be in an abnormal range at some moment in the future or at some time period in the future, that is, the condenser will not be operating at or near the optimal vacuum degree value at some moment in the future or at some time period in the future. At this time, it is necessary to enter the vacuum abnormality warning mode, that is, it is necessary to adjust the operating status of at least part of the components of the cold end system.
[0067] In a specific implementation, the preset vacuum reference value interval is determined by a Gaussian Mixture Model (GMM) and a Support Vector Regression (SVR) model according to the target prediction result of the condenser vacuum degree and the preset vacuum reference value interval.
[0068] When it is determined that it is necessary to enter the vacuum abnormality warning mode, the relevant staff is notified through an alarm or other reminder methods to adjust the operating status of at least part of the components of the cold end system to ensure stable operation of the cold end system.
[0069] It is not difficult to see that the condenser vacuum value prediction method provided in this embodiment only uses the historical values of the target related parameters whose correlation values with the vacuum in the cold end system meet the preset correlation conditions for model training, which reduces the influence of the values of the operating parameters that have little or no influence on the condenser vacuum value on the model for condenser vacuum value prediction, thereby improving the model training efficiency and the prediction accuracy of the trained model. After the condenser vacuum value prediction model (i.e., the preset target condenser vacuum value prediction model) is trained, the target data sequence of multiple parameter values of the target related parameters arranged in the corresponding chronological order is directly input into the model for condenser vacuum value prediction, ensuring the high accuracy and reliability of the condenser vacuum value prediction.
[0070] Figure 1 FIG. 1 is a flow chart of a method for predicting the condenser vacuum value in one embodiment. It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.
[0071] Figure 2 The structure block diagram of the condenser vacuum value prediction device according to the embodiment of the present application is schematically shown. Figure 2 As shown, in one embodiment of the present application, a condenser vacuum value prediction device is provided, and the condenser vacuum value prediction device includes a relevant parameter value acquisition module and a vacuum value prediction module, wherein: The relevant parameter value acquisition module is used to acquire a target data sequence for each target relevant parameter in the target relevant parameter set for the condenser, wherein the target relevant parameter is an operating parameter whose correlation value with the vacuum degree in the cold end system meets a preset correlation condition.
[0072] The vacuum value prediction module is used to input each target data sequence into a preset target condenser vacuum value prediction model to obtain a target prediction result of the condenser vacuum.
[0073] In this embodiment, the target-related parameter set includes circulating water outlet temperature, hot well condensate temperature, low-pressure cylinder exhaust temperature, unit load, condensate flow rate and circulating water inlet temperature.
[0074] In this embodiment, the device further includes: The historical data set acquisition module is used to acquire the historical data set, wherein the historical data set includes multiple historical data subsets; each historical data subset includes the historical values of each operating parameter corresponding to multiple historical moments, and the historical value of the condenser vacuum degree at at least one future moment corresponding to the multiple historical moments.
[0075] The steady-state data determination module is used to perform steady-state screening on a historical data subset in a historical data set based on a preset steady-state condition to obtain multiple steady-state data subsets.
[0076] The model training module is used to use the steady-state data subset to perform model training on the preset initial condenser vacuum value prediction model to obtain the preset target condenser vacuum value prediction model.
[0077] In this embodiment, the device further includes: The correlation value determination module is used to determine the correlation value corresponding to each operating parameter based on the historical value of the operating parameter and the historical value of the condenser vacuum in each steady-state data subset using the spearman correlation coefficient.
[0078] The target-related parameter determination module is used to determine, for each operating parameter, if the correlation value corresponding to the operating parameter meets a preset correlation condition, that the operating parameter is a target-related parameter.
[0079] In this embodiment, the model training module includes: The sample set division unit is used to determine a training sample set and a test sample set according to each steady-state data subset, wherein the training sample set includes a plurality of first steady-state data subsets, and the test sample set includes a plurality of second steady-state data subsets.
[0080] The model training unit is used to train the preset initial condenser vacuum value prediction model for each first steady-state data subset, using the historical values of each target-related parameter in the first steady-state data subset as model input and the historical values of condenser vacuum as model output to obtain a pending model.
[0081] The model prediction unit is used to use the historical values of each target-related parameter in the second steady-state data subset as the input of the prediction model to be evaluated for each second steady-state data subset, so as to obtain the prediction result corresponding to the second steady-state data subset.
[0082] The model determination unit is used to determine whether to use the test sample set to perform model training on the pending model based on the historical values of the condenser vacuum in each second steady-state data subset and the corresponding prediction results. If so, the test sample set is used to perform model training on the pending model to obtain a preset target condenser vacuum value prediction model. If not, the pending model is determined as the preset target condenser vacuum value prediction model.
[0083] In this embodiment, the model determination unit includes: The error determination subunit is used to determine the overall prediction error value of the pending model according to the historical value of the condenser vacuum degree in each second steady-state data subset and the prediction result corresponding to each second steady-state data subset.
[0084] The secondary training judgment subunit is used to determine whether to use the test sample set to perform model training on the model to be determined based on the overall prediction error value and the preset error threshold.
[0085] In this embodiment, the device further includes: The vacuum warning module is used to determine whether it is necessary to enter the vacuum abnormality warning mode based on the target prediction result of the condenser vacuum and the preset vacuum reference value range.
[0086] In this embodiment, the condenser vacuum value prediction device includes a processor and a memory. The relevant parameter value acquisition module and the vacuum value prediction module are stored in the memory as program units, and the processor executes the above program modules stored in the memory to implement corresponding functions.
[0087] The processor includes a kernel, and the kernel retrieves the corresponding program unit from the memory. One or more kernels can be provided, and the above-mentioned condenser vacuum value prediction method is realized by adjusting kernel parameters.
[0088] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.
[0089] An embodiment of the present application provides a processor configured to execute the above-mentioned condenser vacuum value prediction method.
[0090] An embodiment of the present application provides a machine-readable storage medium, on which instructions are stored. When the instructions are executed by a processor, the processor is configured to execute the above-mentioned condenser vacuum value prediction method.
[0091] Figure 3 The internal structure diagram of the computer device of the embodiment of the present application is schematically shown. Figure 3 As shown, in one embodiment of the present application, a computer device is provided, which may be a terminal. The computer device includes a processor A01, a network interface A02, a display screen A04, an input device A05 and a memory (not shown in the figure) connected via a system bus. Among them, the processor A01 of the computer device is used to provide computing and control capabilities. The memory of the computer device includes an internal memory A03 and a non-volatile storage medium A06. The non-volatile storage medium A06 stores an operating system B01 and a computer program B02. The internal memory A03 provides an environment for the operation of the operating system B01 and the computer program B02 in the non-volatile storage medium A06. The network interface A02 of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor A01, a condenser vacuum value prediction method is implemented. The display screen A04 of the computer device may be a liquid crystal display or an electronic ink display, and the input device A05 of the computer device may be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.
[0092] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.
[0093] In one embodiment, the condenser vacuum value prediction device provided in the present application can be implemented in the form of a computer program. The computer program can be Figure 3 The computer device shown in the figure is run. The memory of the computer device can store various program modules constituting the condenser vacuum value prediction device, for example, Figure 2 The relevant parameter value acquisition module and vacuum value prediction module shown in the figure. The computer program composed of various program modules enables the processor to execute the steps of the condenser vacuum value prediction method of each embodiment of the present application described in this specification.
[0094] Figure 3 The computer device shown can be Figure 2 The relevant parameter value acquisition module in the condenser vacuum value prediction device shown executes the step 200 , and the vacuum value prediction module executes the step 400 .
[0095] The present application provides a device, which includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, the following steps are implemented: A target data sequence for each target-related parameter in a target-related parameter set for the condenser is obtained; wherein the target-related parameter is an operating parameter whose correlation value with the vacuum degree in the cold end system meets a preset correlation condition.
[0096] Each target data sequence is input into a preset target condenser vacuum value prediction model to obtain a target prediction result of the condenser vacuum.
[0097] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.
[0098] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0099] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0100] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0101] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.
[0102] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.
[0103] Computer readable media include permanent and non-permanent, removable and non-removable media, and can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.
[0104] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.
[0105] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.
Claims
1. A method for predicting the vacuum value of a condenser, characterized in that: The method comprises: Obtaining a target data sequence for each target-related parameter in a target-related parameter set for the condenser; wherein the target-related parameter is an operating parameter whose correlation value with the vacuum degree in the cold end system satisfies a preset correlation condition; Each target data sequence is input into a preset target condenser vacuum value prediction model to obtain a target prediction result of the condenser vacuum.
2. The method according to claim 1, characterized in that: Before inputting each target data sequence into a preset target condenser vacuum value prediction model, the method further includes: Acquire a historical data set; wherein the historical data set includes a plurality of historical data subsets; each of the historical data subsets includes a historical value of each operating parameter corresponding to a plurality of historical moments, and a historical value of a condenser vacuum degree at least one future moment corresponding to the plurality of historical moments; Based on the preset steady-state condition, a steady-state screening is performed on a historical data subset in the historical data set to obtain multiple steady-state data subsets; The preset initial condenser vacuum value prediction model is trained using the steady-state data subset to obtain the preset target condenser vacuum value prediction model.
3. The method according to claim 2, characterized in that Before using the steady-state data subset to perform model training on the preset initial condenser vacuum value prediction model, the method further includes: For each operating parameter, based on the historical value of the operating parameter and the historical value of the condenser vacuum in each steady-state data subset, the Spearman correlation coefficient is used to determine the correlation value corresponding to the operating parameter; For each operating parameter, if the correlation value corresponding to the operating parameter meets the preset correlation condition, the operating parameter is determined to be a target correlation parameter.
4. The method according to claim 3, characterized in that: The preset initial condenser vacuum value prediction model is trained using the steady-state data subset to obtain the preset target condenser vacuum value prediction model, including: Determine a training sample set and a test sample set according to each steady-state data subset; wherein the training sample set includes a plurality of first steady-state data subsets, and the test sample set includes a plurality of second steady-state data subsets; For each first steady-state data subset, the historical values of each target-related parameter in the first steady-state data subset are used as model input, and the historical values of the condenser vacuum are used as model output, and the preset initial condenser vacuum value prediction model is trained to obtain a pending model; For each second steady-state data subset, the historical values of each target-related parameter in the second steady-state data subset are used as inputs of the prediction model to be evaluated to obtain a prediction result corresponding to the second steady-state data subset; According to the historical values of the condenser vacuum and the corresponding prediction results in each second steady-state data subset, determine whether to use the test sample set to train the model to be determined. If so, use the test sample set to train the model to determine the preset target condenser vacuum value prediction model. If not, determine the model to be determined as the preset target condenser vacuum value prediction model.
5. The method according to claim 4, characterized in that According to the historical values of the condenser vacuum in each second steady-state data subset and the corresponding prediction results, determining whether to use the test sample set to perform model training on the model to be determined includes: Determine the overall prediction error value of the undetermined model according to the historical value of the condenser vacuum degree in each second steady-state data subset and the prediction result corresponding to each second steady-state data subset; According to the overall prediction error value and the preset error threshold, determine whether to use the test sample set to train the model to be determined.
6. The method according to any one of claims 1 to 5, characterized in that After inputting each target data sequence into a preset target condenser vacuum value prediction model to obtain a target prediction result of the condenser vacuum, the method further includes: Based on the target prediction result of the condenser vacuum and the preset vacuum reference value range, determine whether it is necessary to enter the vacuum abnormality warning mode.
7. The method according to any one of claims 1 to 5, characterized in that The target-related parameter set includes circulating water outlet temperature, hot well condensate temperature, low-pressure cylinder exhaust temperature, unit load, condensate flow rate and circulating water inlet temperature.
8. A condenser vacuum value prediction device, characterized in that: The device comprises: A relevant parameter value acquisition module is used to obtain a target data sequence for each target relevant parameter in a target relevant parameter set for the condenser; wherein the target relevant parameter is an operating parameter whose correlation value with the vacuum degree in the cold end system meets a preset correlation condition; The vacuum value prediction module is used to input each target data sequence into a preset target condenser vacuum value prediction model to obtain a target prediction result of the condenser vacuum.
9. A processor, characterized in that: The method is configured to execute the condenser vacuum value prediction method according to any one of claims 1 to 7.
10. A machine-readable storage medium having instructions stored thereon, characterized in that: When the instruction is executed by a processor, the processor is configured to execute the condenser vacuum value prediction method according to any one of claims 1 to 7.