A centrifugal air compressor thermal energy management method and system

Abstract

The present invention relates to the field of artificial intelligence technology, and the present invention relates to a centrifugal air compressor thermal energy management method and system, comprising: first, training a neural network model based on centrifugal air compressor operating parameter data and exhaust port temperature data in historical data to obtain a trained neural network model for predicting exhaust port temperature data; then, classifying the predicted exhaust port temperature data to obtain multiple temperature categories, and searching the historical database for the initial cooling water flow rate, cooling water cooling efficiency, cooling oil flow rate, and cooling oil cooling efficiency corresponding to each temperature category; determining the temperature category to which the real-time exhaust temperature belongs, and dynamically adjusting the flow rates of cooling water and cooling oil according to the temperature deviation between the real-time exhaust temperature and the steady-state temperature to ensure that the system temperature reaches the preset steady-state temperature. The present invention solves the problem of accurate control of the thermal energy of a centrifugal air compressor.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology. More specifically, the present invention relates to a centrifugal air compressor thermal energy management method and system. Background Art

[0002] With the rapid development of industrial technology, centrifugal air compressors, as an efficient compression equipment, are widely used in many industrial fields, including petroleum, chemical, metallurgy, electronics, pharmaceuticals, food processing and energy. Its main function is to provide stable compressed air for various industrial fields to ensure the smooth progress of the production process. In these fields, the operating efficiency and reliability of centrifugal air compressors directly affect the continuity of production and the quality of products, and are therefore highly valued by the industry. Compared with other types of air compressors, centrifugal air compressors have better stability and reliability, ensure the power transmission between mechanical systems, and provide stable power output for other connecting devices outside the equipment. However, centrifugal air compressors will inevitably generate a lot of heat energy during operation. Due to the significant energy conversion in the process of compressing air, most of the electrical energy will be converted into heat energy. If this heat energy cannot be dissipated in time and effectively, it will cause the internal temperature of the air compressor to continue to rise. This temperature increase will not only affect the working efficiency of the air compressor and increase energy consumption, but may also cause overheating damage to the key components of the equipment, thereby shortening the service life of the equipment. Therefore, how to effectively manage and control the thermal energy of centrifugal air compressors during operation has become a key technical issue in industrial applications.

[0003] Traditional centrifugal air compressor thermal management methods mostly set fixed cooling water flow rates and cooling oil flow rates based on design experience to meet the heat dissipation requirements under different working conditions. Although this fixed control method can meet the basic operating requirements of the equipment in the early design stage, with the changes in actual working conditions, especially in the face of complex and changing working environments, this method gradually exposes many shortcomings and it is difficult to achieve accurate control of the thermal energy of the centrifugal air compressor. Summary of the invention

[0004] In order to solve the problem of difficulty in achieving accurate control of the thermal energy of a centrifugal air compressor raised in the above background technology, the present invention provides solutions in the following aspects.

[0005] In a first aspect, the present invention provides a method for thermal management of a centrifugal air compressor, comprising:

[0006] The operating parameter data and exhaust port temperature data of the centrifugal air compressor are collected and input into the neural network model, and the predicted data of the exhaust port temperature are output; the predicted data of the exhaust port temperature are classified to obtain multiple temperature categories; when the exhaust port temperature is T′, the temperature category to which T′ belongs is determined, and the initial cooling water flow rate, initial cooling water cooling efficiency, initial cooling oil flow rate and initial cooling oil cooling efficiency of the temperature category to which T′ belongs are searched from the historical database; the cooling water flow rate and the cooling oil flow rate are adjusted in real time, and the specific adjustments are as follows:

[0007]

[0008] Where W and O represent the adjusted cooling water flow rate and cooling oil flow rate, respectively; a and b represent the adjustment factors; Yi and Zi represent the initial cooling water flow rate and initial cooling oil flow rate corresponding to the i-th temperature category, respectively; Pi and Qi represent the initial cooling water cooling efficiency and initial cooling oil cooling efficiency corresponding to the i-th temperature category, respectively; T c Indicates the preset steady-state temperature.

[0009] The above technical solution improves the accuracy and dynamic response capability of the centrifugal air compressor cooling system by adjusting the cooling water flow rate and the cooling oil flow rate in real time, optimizes energy usage, and achieves precise control of the thermal energy of the centrifugal air compressor.

[0010] Furthermore, the operating parameter data and exhaust port temperature data of the centrifugal air compressor are collected through a sensor network.

[0011] Furthermore, the neural network model is a BP neural network model.

[0012] BP neural network can effectively predict the exhaust port temperature through its powerful nonlinear modeling ability and automatic feature learning function, thus realizing accurate cooling system adjustment. This technology not only improves the prediction accuracy and system intelligence, but also reduces human intervention and enhances the operation stability and overall performance of the equipment.

[0013] Furthermore, the operating parameter data include: primary cooling port exhaust temperature, cooling water flow rate, cooling water outlet temperature, cooling oil flow rate and cooling oil outlet temperature.

[0014] Furthermore, it also includes preprocessing the operating parameter data and the exhaust port temperature data, and the preprocessing process includes: nonlinear normalization, missing value filling and data smoothing.

[0015] The above technical solution can eliminate the scale difference and nonlinear influence of the data by preprocessing the operating parameter data and the exhaust temperature data, and improve the stability, convergence speed and prediction accuracy of the neural network model. This processing method can optimize data input, improve model performance, and improve system efficiency, thereby achieving accurate thermal energy management of centrifugal air compressors.

[0016] Furthermore, a decision tree algorithm or a K-means clustering algorithm is used to classify the predicted data of the exhaust port temperature.

[0017] The K-means algorithm can effectively group data into multiple temperature categories, improving classification accuracy and processing efficiency. The decision tree algorithm can accurately classify the predicted data into multiple temperature categories when classifying the exhaust temperature prediction data, providing an intuitive decision path, and improving the accuracy and interpretability of the classification. The decision tree has strong adaptability, the ability to process large data sets, and the effect of optimizing the adjustment of the cooling system, making it perform well in the thermal energy management of centrifugal air compressors.

[0018] Furthermore, it also includes training a neural network model, specifically:

[0019] The centrifugal air compressor operating parameter data and exhaust port temperature data in the historical data are input into the pre-built neural network model for training. During the training process, the loss between the output prediction value and the label is calculated; the neural network model parameters are adjusted using the gradient descent method to minimize the prediction error; the parameters of the neural network model are iteratively adjusted until the loss is less than a certain value or the set number of training times is reached, and finally a trained neural network model is obtained.

[0020] In a second aspect, the present invention provides a centrifugal air compressor thermal energy management system, comprising a memory and a processor, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a centrifugal air compressor thermal energy management method described in any one of the above is implemented.

[0021] The beneficial effects of the present invention are:

[0022] The present invention significantly improves the accuracy and efficiency of centrifugal air compressor thermal management by combining neural network prediction, decision tree classification, nonlinear normalization processing and real-time adjustment control. Accurate temperature prediction and classification, as well as real-time cooling adjustment, not only optimize the performance of the cooling system, but also save energy and costs. In addition, the system's automated management and data processing improve the model's training effect, making the entire thermal management process more efficient, stable and reliable. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0024] Figure 1 is a flow chart schematically showing a method for thermal energy management of a centrifugal air compressor according to an embodiment of the present invention;

[0025] Figure 2 The figure schematically shows a structural block diagram of a centrifugal air compressor thermal energy management system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0027] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0028] A method for thermal energy management of a centrifugal air compressor is disclosed.

[0029] like Figure 1 As shown, a flow chart of a centrifugal air compressor thermal energy management method according to an embodiment of the present invention comprises the following steps:

[0030] S1: Collect the operating parameter data and exhaust port temperature data of the centrifugal air compressor.

[0031] In one embodiment, the operating parameter data and exhaust port temperature data of the centrifugal air compressor can be collected through a sensor network, and of course, can also be collected through field bus or industrial Ethernet; the operating parameter data include: primary cooling port exhaust temperature, cooling water flow rate, cooling water outlet temperature, cooling oil flow rate and cooling oil outlet temperature.

[0032] Among them, the exhaust temperature of the first-stage cooling port reflects the temperature of the compressed gas after the first-stage compression. If it is too high, it may cause the equipment to overheat, affecting stability and life; if it is too low, it may cause energy efficiency loss or overcooling, wasting energy. The cooling water flow rate directly affects the efficiency of the cooling system. If the flow rate is too low, it will lead to insufficient cooling and overheating of the equipment; if the flow rate is too high, it may reduce the cooling efficiency and increase energy consumption. The cooling water outlet temperature reflects the cooling effect. If the outlet temperature is too high, it indicates that the cooling capacity is insufficient, affecting heat dissipation; if it is too low, it may lead to overcooling and increase cooling costs. If the flow rate of the cooling oil is too low, the oil will not be able to effectively take away the heat generated inside the equipment, which will cause local overheating of the equipment. If the flow rate of the cooling oil is too high, the oil may flow too fast and fail to fully absorb the heat inside the equipment, resulting in reduced cooling effect. The cooling oil outlet temperature reflects the heat dissipation of the cooling oil. If the temperature is too high, it may accelerate oil deterioration and reduce the lubrication effect; if the temperature is too low, it may lead to overcooling and waste energy.

[0033] Specifically, it also includes preprocessing of operating parameter data and exhaust port temperature data, and nonlinear normalization processing to make the data more suitable for machine learning model training by adjusting the data range, processing nonlinear relationships, enhancing model stability and improving training efficiency. This not only improves the prediction accuracy of the model, but also optimizes the training process; missing value filling is used to fill in the missing values ​​in the data set to ensure data integrity; data smoothing is used to reduce noise and irregular fluctuations in the data, making the data smoother and more representative.

[0034] S2: Input the collected real-time operating parameters and exhaust port temperature data into the neural network model to generate predicted exhaust port temperature data.

[0035] In one embodiment, the neural network model can adopt a feedforward neural network model or a BP neural network model. Taking the BP neural network model as an example, training the neural network model includes: obtaining a training set; wherein the training set includes centrifugal air compressor operating parameter data and exhaust port temperature data in historical data; wherein the label is manually set.

[0036] The training set is input into the pre-built neural network model for training. During the training process, the loss between the output prediction value and the label is calculated; the neural network model parameters are adjusted using the gradient descent method to minimize the prediction error; the parameters of the neural network model are iteratively adjusted until the loss is less than a certain value or the set number of training times is reached, and finally a trained neural network model is obtained.

[0037] Finally, the real-time collected centrifugal air compressor operating parameter data and exhaust port temperature data are input into the trained neural network model to output the predicted exhaust port temperature data.

[0038] S3: Classifying the predicted data of the exhaust port temperature into a plurality of temperature categories, and extracting parameter data corresponding to each temperature category from a historical database.

[0039] In one embodiment, the predicted exhaust port temperature data may be classified using a decision tree algorithm, and of course, a random forest algorithm may also be used for classification; the predicted exhaust port temperature data is divided into a plurality of temperature categories.

[0040] Assume that the predicted exhaust temperature data set is {105, 107, 115, 117, 121, 129}, unit: °C; use the decision tree algorithm to classify and obtain the following different temperature categories: temperature category 1 (100 °C-110 °C): 105, 107; temperature category 2 (111 °C-120 °C): 115, 117; temperature category 3 (121 °C-130 °C): 121, 129; and search the historical database for the initial cooling water flow rate, initial cooling water cooling efficiency, initial cooling oil flow rate and initial cooling oil cooling efficiency corresponding to each temperature category.

[0041] By using a decision tree algorithm to classify the predicted exhaust temperature data, the exhaust temperature can be accurately assigned to the corresponding category, thereby achieving precise cooling water and cooling oil flow rate adjustment, optimizing the thermal energy management of the centrifugal air compressor, and improving equipment operation efficiency and safety.

[0042] S4: Determine the temperature category to which the real-time exhaust port temperature belongs, and dynamically adjust the cooling water flow rate and the cooling oil flow rate.

[0043] In one embodiment, when the real-time exhaust port temperature is T′, the temperature category to which T′ belongs is determined, and the cooling water flow rate and the cooling oil flow rate are adjusted in real time. The specific adjustments are as follows:

[0044]

[0045] Where W and O represent the adjusted cooling water flow rate and cooling oil flow rate, respectively; a and b represent the adjustment factors; Yi and Zi represent the initial cooling water flow rate and initial cooling oil flow rate corresponding to the i-th temperature category, respectively; Pi and Qi represent the initial cooling water cooling efficiency and initial cooling oil cooling efficiency corresponding to the i-th temperature category, respectively; T c Indicates the preset steady-state temperature.

[0046] Assuming that the real-time exhaust port temperature T′ is 125°C and the preset steady-state temperature is 120°C, it can be seen from the above classification results that the exhaust port temperature is temperature category 3 at this time, and the initial cooling water flow rate corresponding to temperature category 3 is 60L / min, the initial cooling water cooling efficiency is 0.85, the initial cooling oil flow rate is 35L / min, and the initial cooling oil cooling efficiency is 0.7, which is found from the historical database; assuming that the value of the adjustment factor a is 3 and the value of the adjustment factor b is 2, substitute them into the above formula to obtain the adjusted cooling water flow rate of 62.55 / min and the adjusted cooling oil flow rate of 36.4L / min.

[0047] The above formula is used to design appropriate cooling water flow rate and cooling oil flow rate. The adjustment factors of cooling water and cooling oil can be changed according to actual conditions or emergencies. For example, when a cooling water system failure occurs, the adjustment factor b can be increased to keep the air compressor in a stable working state. The algorithm improves the flexibility and robustness of the system. Combined with the BP neural network prediction, negative feedback regulation of the temperature in the system can be achieved to avoid high temperature failures of centrifugal air compressors.

[0048] The above technical solution collects the operating parameter data and exhaust port temperature data of the centrifugal air compressor and inputs them into the neural network model for prediction, thereby providing an accurate exhaust port temperature prediction value. Classifying these prediction data to obtain different temperature categories helps to formulate targeted cooling strategies. This method can adjust the cooling water flow rate and cooling oil flow rate in real time to maintain the equipment within the optimal operating temperature range, thereby improving the thermal energy management efficiency of the air compressor.

[0049] An embodiment of a centrifugal air compressor thermal energy management system:

[0050] like Figure 2 As shown, a structural block diagram of a centrifugal air compressor thermal energy management system according to an embodiment of the present invention includes a processor and a memory.

[0051] The present invention also provides a communication anomaly detection system for a wind farm. Figure 2 As shown, the system includes a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, any one of the above-mentioned centrifugal air compressor thermal energy management methods is implemented.

[0052] The centrifugal air compressor thermal energy management system also includes other components familiar to those skilled in the art, such as a communication interface, and their configuration and functions are known in the art, so they will not be described in detail here.

[0053] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, apparatus or device. For example, a computer-readable storage medium may be any suitable magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM (Resistive Random Access Memory), a dynamic random access memory DRAM (Dynamic Random Access Memory), a static random access memory SRAM (Static Random-Access Memory), an enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), a high-bandwidth memory HBM (High-Bandwidth Memory), a hybrid memory cube HMC (Hybrid Memory Cube), etc., or any other medium that may be used to store the required information and may be accessed by an application, a module, or both. Any such computer storage medium may be part of a device or accessible or connectable to a device. Any application or module described in the present invention may be implemented using computer-readable / executable instructions that may be stored or otherwise maintained by such a computer-readable medium.

[0054] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0055] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. A method for thermal energy management of a centrifugal air compressor, characterized in that: include: Collect the operating parameter data and exhaust port temperature data of the centrifugal air compressor and input them into the neural network model, and output the predicted data of the exhaust port temperature; The predicted data of the exhaust port temperature are classified to obtain a plurality of temperature categories; when the exhaust port temperature is T ′ When T ′ The temperature category to which it belongs is searched from the historical database. ′ The initial cooling water flow rate, initial cooling water cooling efficiency, initial cooling oil flow rate and initial cooling oil cooling efficiency of the temperature category; the cooling water flow rate and cooling oil flow rate are adjusted in real time, and the specific adjustments are as follows: Where W and O represent the adjusted cooling water flow rate and cooling oil flow rate, respectively; a and b represent the adjustment factors; Yi and Zi represent the initial cooling water flow rate and initial cooling oil flow rate corresponding to the i-th temperature category, respectively; Pi and Qi represent the initial cooling water cooling efficiency and initial cooling oil cooling efficiency corresponding to the i-th temperature category, respectively; T c Indicates the preset steady-state temperature.

2. A centrifugal air compressor thermal energy management method according to claim 1, characterized in that: The operating parameter data and exhaust port temperature data of the centrifugal air compressor are collected through a sensor network.

3. A centrifugal air compressor thermal energy management method according to claim 1, characterized in that: The neural network model is a BP neural network model.

4. A centrifugal air compressor thermal energy management method according to claim 1, characterized in that: The operating parameter data include: primary cooling port exhaust temperature, cooling water flow rate, cooling water outlet temperature, cooling oil flow rate and cooling oil outlet temperature.

5. The method for thermal energy management of a centrifugal air compressor according to claim 1, characterized in that: It also includes preprocessing the operating parameter data and the exhaust port temperature data, and the preprocessing process includes: nonlinear normalization, missing value filling and data smoothing.

6. A centrifugal air compressor thermal energy management method according to claim 1, characterized in that: The prediction data of the exhaust port temperature is classified by using a decision tree algorithm or a K-means clustering algorithm.

7. A centrifugal air compressor thermal energy management method according to claim 1, characterized in that: It also includes training the neural network model, specifically: The centrifugal air compressor operating parameter data and exhaust port temperature data in the historical data are input into the pre-built neural network model for training. During the training process, the loss between the output prediction value and the label is calculated; Use gradient descent to adjust neural network model parameters to minimize prediction error; Iteratively adjust the parameters of the neural network model until the loss is less than a certain value or the set number of training times is reached, and finally a trained neural network model is obtained.

8. A centrifugal air compressor thermal energy management system, characterized in that: The invention comprises a memory and a processor, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a centrifugal air compressor thermal energy management method as claimed in any one of claims 1 to 7 is implemented.

Images