Federal learning-based perception system evaluation optimization method and system, medium, product and terminal

Abstract

The invention provides a sensing system evaluation optimization method and system based on federated learning, a medium, a product and a terminal, and the method comprehensively considers a plurality of key indexes, such as sensor accuracy, data transmission stability, data integrity, equipment failure rate and maintenance frequency, so as to guarantee the comprehensive evaluation of a refrigerating machine room sensing system. According to the method, through normalization processing and reasonable weight coefficient determination, the evaluation result is more accurate, and the real operation condition of the sensing system can be reflected more accurately; potential problems in the system can be found in time according to the evaluation result, and feedback is provided in the form of early warning information to guide maintenance and optimization strategies, so that the reliability and operation efficiency of the system are improved; according to the method, the federal learning method is adopted to realize practical application in cross-geographic-position sensing system evaluation optimization, so that the overall performance and reliability of the sensing system are effectively improved, and meanwhile, the data privacy is protected.

Description

Technical Field

[0001] The present application relates to the field of refrigeration technology, and in particular to a perception system evaluation and optimization method, system, medium, product and terminal based on federated learning. Background Art

[0002] In modern refrigeration rooms, the perception system plays a vital role, responsible for monitoring and managing the operation of the entire refrigeration process. The perception system consists of multiple sensors, data transmission modules, and data processing units. It ensures the normal operation and efficient energy management of the perception system in the refrigeration room by collecting and transmitting various operating parameters (such as temperature, pressure, flow, etc.) in real time. However, with the continuous development of refrigeration equipment and technology, the complexity and data volume of the perception system are also increasing, and the assessment and maintenance of its health are becoming more and more important.

[0003] At present, common perception system health assessment methods mainly include the following:

[0004] (1) Sensor performance monitoring: mainly focuses on the accuracy and response speed of the sensor. Ensure its accuracy and reliability by regularly calibrating and checking the working status of the sensor.

[0005] (2) Data transmission quality monitoring: The stability of data transmission is evaluated by monitoring indicators such as packet loss rate and delay during data transmission. This method is mainly used to ensure that data is not lost or damaged during transmission.

[0006] (3) Equipment failure rate statistics: Evaluate the reliability of the sensing system based on historical failure records and failure rate statistics. This method mainly relies on past failure data to predict possible future failures.

[0007] Although the above methods can reflect the partial health status of the perception system to a certain extent, they have the following shortcomings in practical applications:

[0008] (1) Singleness: Most existing methods monitor a single indicator and cannot fully reflect the health status of the entire perception system. For example, sensor performance monitoring only focuses on the accuracy of the sensor, but ignores the stability of data transmission and data integrity.

[0009] (2) Lag: Some evaluation methods rely on historical data and are unable to reflect the real-time health status of the system in a timely manner. For example, equipment failure rate statistics are mainly based on past data and cannot dynamically monitor the current operating status.

[0010] (3) Lack of comprehensive evaluation: Currently, there is no method that can comprehensively consider multiple key indicators such as sensor accuracy, data transmission stability, data integrity, and equipment failure rate to comprehensively evaluate the health of the perception system.

[0011] (4) Lack of early warning capability: Most existing methods lack the ability to warn of potential problems, making it difficult to provide effective guidance for preventive maintenance. This may result in failure to detect and handle faults in a timely manner, affecting the operating efficiency and reliability of the system. Summary of the invention

[0012] In view of the above-mentioned shortcomings of the prior art, the present invention provides a perception system evaluation and optimization method, system, medium, product and terminal based on federated learning, which is used to solve the problems in the prior art that the evaluation of the refrigeration room perception system is mainly concentrated on a single indicator, which can only reflect the partial health status of the perception system and lacks a comprehensive evaluation of the overall health status of the perception system.

[0013] To achieve the above-mentioned purpose and other related purposes, the first aspect of the present application provides a perception system evaluation and optimization method based on federated learning, including: real-time collection of operating data of each geographical location of the refrigeration room perception system, and preprocessing of the operating data; constructing and training based on the preprocessed operating data of each geographical location to obtain a health assessment sub-model corresponding to each geographical location; based on the health assessment sub-model corresponding to each geographical location, using the secure multi-party computing technology in the federated learning method to perform parameter aggregation to obtain a health assessment global model; performing calculations based on the health assessment global model to evaluate the health of the refrigeration room perception system; and optimizing and adjusting the refrigeration room perception system according to the evaluation results of the health of the refrigeration room perception system.

[0014] In some embodiments of the first aspect of the present application, the health assessment sub-model is constructed by several key health indicators; the key health indicators include sensor accuracy, data transmission stability, data integrity, equipment failure rate and maintenance frequency.

[0015] In some embodiments of the first aspect of the present application, each of the key health indicators is normalized and the weight coefficient is determined to perform a weighted sum calculation; the calculation formula of each health assessment sub-model is: SH=k1×SA+k2×DS+k3×DI+k4×ER+k5×MF; wherein SA is the normalized sensor accuracy, DS is the normalized data transmission stability, DI is the normalized data integrity, ER is the normalized equipment failure rate, MF is the normalized maintenance frequency, and k1, k2, k3, k4, and k5 are the corresponding weight coefficients respectively.

[0016] In some embodiments of the first aspect of the present application, the method further comprises: the calculation formula of the normalized sensor accuracy is: SA = D C / D A; where SA is the normalized sensor accuracy, D C is the correct number of data items, D T is the number of data actually monitored; the calculation formula for the normalized data transmission stability is: DS = D A / D N ; where DS is the normalized data transmission stability, D T is the number of data actually monitored, D N is the number of standard data transmissions; the calculation formula for the normalized data integrity is: DI = D I / D A ; Among them, DI is the normalized data integrity, D I is the number of complete data items, D A The actual number of monitored data.

[0017] In some embodiments of the first aspect of the present application, the method further comprises: the calculation formula of the normalized equipment failure rate is: ER = F A / F N ; Among them, ER is the normalized equipment failure rate, F A is the actual failure rate, F N is the standard failure rate; the calculation formula of the normalized maintenance frequency is: MF = M A / M N ; Among them, MF is the normalized maintenance frequency, M A is the actual number of maintenance times, M N This is the standard maintenance frequency.

[0018] In some embodiments of the first aspect of the present application, the method further includes: sending corresponding optimization adjustment suggestions to each geographical location of the refrigeration room sensing system according to the evaluation result of the health of the refrigeration room sensing system.

[0019] To achieve the above-mentioned purpose and other related purposes, the second aspect of the present application provides a perception system evaluation and optimization system based on federated learning, including: an operation data acquisition module, which is used to collect the operation data of each geographical location of the refrigeration room perception system in real time and pre-process the operation data; a sub-model construction module, which is used to construct and train the pre-processed operation data of each geographical location respectively, so as to obtain the health assessment sub-model corresponding to each geographical location; a global model acquisition module, which is used to perform parameter aggregation based on the health assessment sub-model corresponding to each geographical location using the secure multi-party computing technology in the federated learning method to obtain a health assessment global model; a health calculation and evaluation module, which is used to perform calculations based on the health assessment global model to evaluate the health of the refrigeration room perception system; an optimization and adjustment module, which is used to optimize and adjust the refrigeration room perception system according to the evaluation result of the health of the refrigeration room perception system.

[0020] To achieve the above-mentioned purpose and other related purposes, the third aspect of the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the perception system evaluation and optimization method based on federated learning.

[0021] To achieve the above-mentioned purpose and other related purposes, the fourth aspect of the present application provides a computer program product, which includes a computer program code. When the computer program code runs on a computer, the computer implements the perception system evaluation and optimization method based on federated learning.

[0022] To achieve the above-mentioned purpose and other related purposes, the fifth aspect of the present application provides an electronic terminal, including a memory, a processor and a computer program stored in the memory; the processor executes the computer program to implement the perception system evaluation and optimization method based on federated learning.

[0023] As described above, the perception system evaluation and optimization method, system, medium, product and terminal based on federated learning provided by this application have the following beneficial effects:

[0024] (1) The present invention comprehensively considers multiple key indicators such as sensor accuracy, data transmission stability, data integrity, equipment failure rate, and maintenance frequency to ensure a comprehensive evaluation of the refrigeration room sensing system.

[0025] (2) The present invention makes the evaluation result more accurate through normalization processing and reasonable weight coefficient determination, which helps to more accurately reflect the actual operating status of the perception system.

[0026] (3) The present invention can timely discover potential problems in the system based on the evaluation results, and provide feedback in the form of early warning information to guide maintenance and optimization strategies, thereby improving the reliability and operating efficiency of the system.

[0027] (4) The present invention adopts the federated learning method to achieve practical application in the evaluation and optimization of perception systems across geographical locations, effectively improving the overall performance and reliability of the perception system while protecting data privacy. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 Shown is a flow chart of a perception system evaluation and optimization method based on federated learning in one embodiment of the present application.

[0029] Figure 2 Shown is a specific embodiment diagram of a perception system evaluation and optimization method based on federated learning in one embodiment of the present application.

[0030] Figure 3 Shown is a structural diagram of a perception system evaluation and optimization system based on federated learning in one embodiment of the present application.

[0031] Figure 4 Shown is a schematic diagram of the structure of an electronic terminal in one embodiment of the present application. DETAILED DESCRIPTION

[0032] The following describes the embodiments of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0033] Before further describing the present invention in detail, the nouns and terms involved in the embodiments of the present invention are explained. The nouns and terms involved in the embodiments of the present invention are applicable to the following interpretations:

[0034] <1> Federated Learning is an emerging machine learning technology that aims to achieve joint training of multi-party data while protecting data privacy. Unlike traditional centralized machine learning methods, federated learning allows multiple participants (such as devices or organizations in different geographical locations) to calculate model updates locally and transmit the updated parameters to a central server for aggregation, rather than centralizing the original data in one place. This approach can both improve data utilization and effectively protect data privacy.

[0035] <2> Secure Multi-Party Computation (SMPC) is a cryptographic technology that allows multiple parties to jointly complete a computing task while protecting the privacy of their inputs. In other words, even if the parties do not trust each other, SMPC ensures that they can collaborate securely without leaking their private data.

[0036] An embodiment of the present invention provides a health assessment method for a refrigeration room perception system. By considering multiple key indicators such as comprehensive sensor accuracy, data transmission stability, data integrity, etc., and combining the practical application of cross-geographical refrigeration room perception system health assessment and optimization using a federated learning method, the health status of the perception system is quantitatively assessed, thereby providing a scientific basis for the preventive maintenance and optimized operation of the refrigeration room perception system.

[0037] It should be understood that the perception system evaluation and optimization method based on federated learning in the embodiment of the present application can be applied to controllers such as ARM (Advanced RISC Machines), FPGA (Field Programmable Gate Array), SoC (System on Chip), DSP (Digital Signal Processing) or MCU (Microcontroller Unit); it can also be applied to personal computers such as desktop computers, laptops, tablets, smart phones, smart bracelets, smart watches, smart helmets, smart TVs, personal digital assistants (PDA for short); it can also be applied to servers, and the servers can be arranged on one or more physical servers according to various factors such as function and load, or they can be composed of distributed or centralized server clusters, which is not limited in this embodiment.

[0038] To facilitate understanding of the embodiments of the present application, first Figure 1 Detailed description. Figure 1 The flowchart of a perception system evaluation and optimization method based on federated learning in an embodiment of the present invention is shown. The perception system evaluation and optimization method based on federated learning in this embodiment mainly includes the following steps:

[0039] Step S11: collecting the operating data of each geographical location of the refrigeration room sensing system in real time, and preprocessing the operating data.

[0040] It should be noted that this embodiment can distribute the task of evaluating the health of the perception system to multiple geographical locations or multiple devices for processing through federated learning, while protecting data privacy, and can also use global data to optimize the overall performance and reliability of the perception system. The federated learning method allows multiple participants to collaborate in training machine learning models while maintaining data privacy and localization. The key technologies involved include:

[0041] (1) Federated Averaging: This is the most commonly used algorithm in federated learning. Each participant trains the model locally and transmits the model parameters to the central server. The central server performs weighted averaging on these parameters, updates the global model, and distributes the updated model parameters back to each participant.

[0042] (2) Secure multi-party computation: Ensure that private information is not leaked when computing between multiple parties. In federated learning, SMPC is used to securely aggregate the model parameters or gradients of the participants.

[0043] (3) Differential Privacy: Protecting individual data privacy by introducing noise or disturbance in the calculation process. In federated learning, differential privacy can be applied to local model training or parameter aggregation to ensure data privacy.

[0044] (4) Model compression and quantization: In order to reduce communication and computing costs, model compression and quantization techniques are widely used in federated learning to reduce the amount of transmitted data and the complexity of calculation.

[0045] The main advantages of using a federated learning approach include the following:

[0046] (1) Data privacy protection: Data does not leave the local server, reducing the risk of data leakage.

[0047] (2) Distributed computing: Utilize the computing resources of each participant to reduce the burden on the central server.

[0048] (3) Improve model performance: Improve the generalization ability and accuracy of the model by training with data from multiple parties.

[0049] Specifically, the refrigeration room is divided into several geographical locations, and the operation data of each geographical location is collected in real time, and the operation data is preprocessed. The operation data mainly includes sensor accuracy, data transmission status, equipment failure records, etc. The collected operation data is cleaned and normalized to ensure the accuracy and consistency of the data. Among them, data cleaning can supplement the missing parts of the data, correct or delete the incorrect parts of the data, filter and remove the duplicate and redundant parts of the data, and finally organize the data into data that is easy to analyze and use. The purpose of preprocessing is to improve data quality and ensure the accuracy and integrity of the data.

[0050] Step S12: construct and train the preprocessed operation data of each geographical location to obtain the health assessment sub-model corresponding to each geographical location. The health assessment sub-model is constructed by a number of key health indicators; the key health indicators include sensor accuracy, data transmission stability, data integrity, equipment failure rate and maintenance frequency.

[0051] In some examples, each of the key health indicators is normalized and a weight coefficient is determined to perform a weighted sum calculation; the calculation formula of each health assessment sub-model is:

[0052] SH=k1×SA+k2×DS+k3×DI+k4×ER+k5×MF; (Formula 1)

[0053] Among them, SA is the normalized sensor accuracy, DS is the normalized data transmission stability, DI is the normalized data integrity, ER is the normalized equipment failure rate, MF is the normalized maintenance frequency, and k1, k2, k3, k4, and k5 are the corresponding weight coefficients. k1 is the weight coefficient corresponding to the normalized sensor accuracy SA, k2 is the weight coefficient corresponding to the normalized data transmission stability DS, k3 is the weight coefficient corresponding to the normalized data integrity DI, k4 is the weight coefficient corresponding to the normalized equipment failure rate ER, and k5 is the weight coefficient corresponding to the normalized maintenance frequency MF. In federated learning, each geographic location uses the key indicator features of sensor accuracy, data transmission stability, data integrity, equipment failure rate, and maintenance frequency to train and obtain the health assessment sub-model corresponding to each geographic location.

[0054] It should be noted that the determination of the weight coefficients k1, k2, k3, k4, and k5 is based on the key indicators and specific application scenarios of the refrigeration room sensing system in actual operation. In this implementation, the weight coefficients corresponding to each key indicator can be determined by the following steps:

[0055] (1) Expert experience method: Collect the opinions of experts in related fields and preliminarily determine the relative importance of each key indicator based on the actual operation experience of the perception system. Experts usually assign weights to each key indicator based on their professional judgment and historical experience.

[0056] (2) Hierarchical analysis method: By constructing a key indicator hierarchy model, the factors affecting the health of the perception system are decomposed into several levels to form a hierarchical structure. The judgment matrix is ​​constructed using the paired comparison method to calculate the weight of each key indicator. At the same time, the rationality of this judgment matrix is ​​ensured through consistency testing.

[0057] (3) Data-driven method: Utilizing a large amount of historical operating data of the refrigeration room sensing system, statistical analysis and machine learning algorithms are used to analyze the impact of key indicators on the health of the sensing system. This method can reveal patterns and relationships in the data, thereby dynamically adjusting the weight coefficient to make it closer to the actual situation of the refrigeration room sensing system.

[0058] Finally, the determination of the weight coefficients k1, k2, k3, k4, and k5 combines expert experience and data analysis results to ensure the scientificity and rationality of the evaluation method. For example, the initial weight coefficients k1, k2, k3, k4, and k5 of the refrigeration room sensing system in this embodiment are determined according to the above steps and are set to 0.2, 0.2, 0.2, 0.3, and 0.1, respectively.

[0059] Furthermore, in this embodiment, the key indicators for constructing the health assessment sub-model include sensor accuracy, data transmission stability, data integrity, equipment failure rate, and maintenance frequency. Each key indicator is calculated based on the operating data of the perception system.

[0060] In one embodiment, the accuracy of the sensor in the refrigeration room sensing system is one of the key indicators for measuring sensor performance. It can reflect the accuracy of data transmission of the sensor in actual applications. At the same time, the higher the sensor accuracy, the more the data truly reflects the actual situation. Considering this indicator can improve the credibility of data analysis results and the accuracy of decision-making, thereby effectively supporting business operations and optimization. Through the quantitative embodiment of sensor accuracy, the calculation formula of the normalized sensor accuracy is:

[0061] SA=D C / D A ; (Formula 2)

[0062] Where SA is the normalized sensor accuracy, D C is the correct number of data items, D T The actual number of monitored data.

[0063] For example, suppose that for a sensor in a certain period of time (such as 100 minutes), one piece of data is recorded every minute. In theory, 100 pieces of data should be actually monitored. The data accuracy of the sensor in this period of time is calculated. That is, if 99 of the 100 pieces of data actually monitored are accurate, then the accuracy of the sensor is 99 / 100. At the same time, when calculating the accuracy of the perception system sensor, the accuracy of each sensor can be taken as a separate data point, and then the average of these data points can be calculated to obtain the sensor accuracy of the entire perception system.

[0064] In one embodiment, the data transmission stability index reflects the continuity and consistency of the data collection process of the refrigeration room perception system. The high stability of the data means that the data collection process is stable and reliable, and is not easily interfered by external factors. This is very important for long-term data monitoring and analysis, and can provide more accurate and coherent data support. Among them, the standard data transmission stability refers to the data transmission stability under ideal conditions (for example, no packet loss rate), and the current actual transmission stability is the actual monitored data transmission stability (for example, packet loss rate). The data transmission stability of the perception system is quantified, and the calculation formula for the normalized data transmission stability is:

[0065] DS=D A / D N ; (Formula 3)

[0066] Among them, DS is the normalized data transmission stability, D T is the number of data actually monitored, D N The number of standard data transmissions.

[0067] For example, if in a certain period of time (such as 100 minutes), theoretically 100 pieces of data should be transmitted from the entire sensing system, but in reality only 80 pieces of data are transmitted, then the data transmission stability in this period is 80 / 100. The larger the value, the more stable the data transmission.

[0068] In one embodiment, data integrity refers to the integrity of data in the actual monitored data. For example, there may be a situation where data is partially lost, resulting in incomplete data although the data is transmitted, that is, normal data integrity is the data integrity under ideal conditions (such as no data loss), and current data integrity is the data integrity actually monitored (for example, data loss rate). The current data integrity in this embodiment is quantified, and the calculation formula for the normalized data integrity is:

[0069] DI=D I / D A ; (Formula 4)

[0070] Among them, DI is the normalized data integrity, D I is the number of complete data items, D A The actual number of monitored data.

[0071] For example, within a certain period of time (such as 100 minutes), complete data entries (such as temperature, pressure, flow) should be recorded every minute. If 100 data items are actually monitored and transmitted, but only 80 of them are complete, the data integrity is 80 / 100. The larger the value, the higher the data integrity.

[0072] Specifically, the above key indicators sensor accuracy, data transmission stability and data integrity are further explained to facilitate distinction.

[0073] Sensor accuracy: For example, suppose that the data of the refrigeration room sensing system are monitored within a certain period of time (such as 100 minutes). The results include 99 out of 100 temperature sensor data are accurate; 98 out of 100 pressure sensor data are accurate; and 97 out of 100 flow sensor data are accurate.

[0074] The calculation results of sensor accuracy are as follows: the accuracy of the temperature sensor is 99 / 100=0.99; the accuracy of the pressure sensor is 98 / 100=0.98; the accuracy of the flow sensor is 97 / 100=0.97; and then the accuracy of the three sensors is averaged respectively, so the system sensor accuracy is (0.99+0.98+0.97) / 3=0.98.

[0075] Data transmission stability: For example, suppose 100 pieces of data should be transmitted within 100 minutes, but only 80 pieces of data are actually transmitted. Therefore, the calculation of data transmission stability is data transmission stability = 80 / 100 = 0.80.

[0076] Data integrity: For example, suppose that ideally 100 complete data should be recorded within 100 minutes (temperature, pressure, flow are not missing), but only 60 of the 100 data actually monitored are complete. Therefore, the data integrity is calculated as data integrity = 60 / 100 = 0.60.

[0077] It should be noted that through the above calculation method, the sensor accuracy, data transmission stability and data integrity of the refrigeration room perception system within a preset time period can be accurately evaluated based on the calculation results, and then the health status of the perception system can be fully understood and evaluated.

[0078] In one embodiment, the calculation formula of the normalized device failure rate is:

[0079] ER=F A / F N ; (Formula 5)

[0080] Among them, ER is the normalized equipment failure rate, F A is the actual failure rate, F N is the standard failure rate; the equipment failure rate is the failure condition of the actual running equipment in the perception system; the standard failure rate is the failure rate of the equipment related to the perception system in the refrigeration room under normal working conditions, that is, the ideal failure rate, and the actual failure rate is the actual failure rate of the equipment in a specific time period. The ratio of the actual failure rate to the standard failure rate in the preset time period is calculated to obtain the normalized equipment failure rate of the perception system. The larger the value, the lower the equipment failure rate.

[0081] In one embodiment, the calculation formula for the normalized maintenance frequency is:

[0082] MF=M A / M N ; (Formula 6)

[0083] Among them, MF is the normalized maintenance frequency, M A is the actual number of maintenance times, M N is the standard maintenance frequency. The standard maintenance frequency refers to the ideal maintenance frequency of the perception system within the preset time period, and the actual maintenance frequency refers to the actual maintenance frequency of the perception system within the preset time period. By calculating the ratio between the actual maintenance frequency and the standard maintenance frequency, the normalized maintenance frequency of the refrigeration room perception system is obtained. The larger the ratio, the higher the maintenance frequency and the healthier the perception system. At the same time, the normalized value range is between 0 and 1. The larger the value, the more times the equipment is maintained.

[0084] Step S13: Based on the health assessment sub-models corresponding to each geographical location, secure multi-party computing technology in the federated learning method is used to perform parameter aggregation to obtain a global health assessment model.

[0085] It should be noted that the specific process of constructing a global model for health assessment of the perception system in combination with the federated learning method in this embodiment includes: first constructing a federated learning architecture, the federated learning architecture includes a central server and several participants, the several participants are the various geographical locations of the refrigeration room perception system, and the data center at each geographical location collects and preprocesses the sensor data of the refrigeration room perception system to ensure the accuracy and consistency of the data; then each geographical location independently uses local operating data to train the perception system health assessment sub-model, that is, each health assessment sub-model calculates the health score of the refrigeration room perception system based on a determined evaluation formula, and finally obtains the health assessment sub-models corresponding to each geographical location; further using the secure multi-party computing technology of federated learning, the parameters of the health assessment sub-models corresponding to each geographical location are transmitted to the central server for secure aggregation, and the central server performs weighted average on the received parameters to update the global health assessment model.

[0086] Step S14: performing calculation based on the health evaluation global model to evaluate the health of the refrigeration room sensing system.

[0087] It should be noted that the operation data collected by monitoring sensors or other detection equipment is calculated based on the global model of health assessment, and the calculated health value of the perception system is compared with the preset health threshold to judge and evaluate the health status of the current perception system. If the health value is lower than the preset health threshold, it means that there is a potential problem or failure, and the early warning mechanism can be automatically triggered to generate early warning information. The early warning information can be an alarm, notification or report, the purpose of which is to remind users or system administrators to pay attention to potential problems.

[0088] Step S15: Optimizing and adjusting the refrigeration room perception system according to the evaluation result of the health of the refrigeration room perception system. According to the evaluation result of the health of the refrigeration room perception system, corresponding optimization and adjustment suggestions are sent to each geographical location of the refrigeration room perception system.

[0089] Specifically, at this time, based on the evaluation results of the health of the refrigeration room perception system, a series of optimization suggestions and adjustment strategies are provided for the perception system, including adjustment of operating parameters, optimization of maintenance plans, and equipment inspection and maintenance. Among them, adjustment of operating parameters includes temperature, pressure, speed, etc. to optimize system performance. Optimization of maintenance plans includes adjustment of maintenance cycles and contents based on the health status of the perception system and predicted maintenance needs. Equipment inspection and maintenance includes guiding users to conduct regular or targeted equipment inspection and maintenance work. By optimizing and adjusting the perception system of the refrigeration room, the reliability of the perception system can be improved, unexpected downtime can be reduced, and overall operating efficiency can be improved.

[0090] Furthermore, the specific optimization process of optimizing and adjusting the refrigeration room sensing system includes:

[0091] (1) Optimization goals: Based on the health assessment results, set optimization goals, such as minimizing energy consumption, maximizing cooling output, and extending service life.

[0092] (2) Optimization strategy: Through the health assessment results, determine the current operating status and potential problems of the refrigeration room, identify the factors affecting efficiency and performance, and formulate an optimization strategy. When formulating the optimization strategy, use optimization algorithms such as genetic algorithms and particle swarm optimization to find the optimal operating parameters and maintenance plan.

[0093] (3) Dynamic adjustment: Monitor the operating data of the refrigeration room in real time and dynamically adjust the operating parameters based on the health assessment results. Combined with the prediction model, it can warn of potential failures in advance and adjust the optimization strategy.

[0094] It should be explained that in this embodiment, the overall health of the refrigeration room perception system can be evaluated through the central server's perception system health assessment global model in the federated learning architecture. Based on the health assessment results, the central server sends optimization adjustment suggestions to each geographic location, such as adjusting the operating mode of the refrigeration room perception system, optimizing equipment maintenance plans, etc.

[0095] Furthermore, the evaluation and optimization results of the perception system based on federated learning and its changing trends can be displayed through a visual interface, which can help users to intuitively understand the comprehensive health status of the perception system of the refrigeration room in real time. Specifically, data visualization technology, such as dashboards and charts, is used to display the health evaluation results, so that users can quickly and intuitively view the real-time status and historical data of the perception system of the refrigeration room. Users can also monitor and manage the operating status of the perception system of the refrigeration room more effectively, discover and solve problems in a timely manner, and improve the reliability and efficiency of the perception system.

[0096] It should be emphasized that the present invention applies the federated learning method to the evaluation and optimization of the perception system based on federated learning, realizes the practical application in the evaluation and optimization of the perception system across geographical locations, effectively improves the overall performance and reliability of the perception system, and protects data privacy at the same time. The specific advantages are as follows:

[0097] (1) Data privacy protection: Federated learning allows data to remain locally, avoiding the cross-geographical transmission of sensitive data and reducing the risk of data leakage. At the same time, by only sharing model parameters or gradient updates in a secure environment, data privacy and security are protected.

[0098] (2) Global optimization: Use global data to optimize the overall performance of the refrigeration room sensing system, including improving energy efficiency, stability, and equipment life.

[0099] (3) Dynamic adjustment and real-time response: Based on real-time data and feedback from the global model, the optimization strategy and operating parameters can be adjusted in a timely manner to achieve real-time monitoring and response of the perception system, thereby improving the response speed and adaptability of the perception system.

[0100] In order to demonstrate the implementation of the federated learning method in this application to achieve the health assessment and optimization of the perception system of the refrigeration room across geographical locations, combined with Figure 2 The following specific examples are provided for illustration:

[0101] Example 1: Assume that the various devices of a large refrigeration room perception system are distributed in multiple geographical locations. The geographical locations may include equipment such as refrigeration machines, cooling towers, and cooling pumps. Multiple sensors are installed on each device to monitor parameters such as temperature, flow, and pressure. Federated learning technology is used to achieve cross-geographic perception system evaluation optimization while protecting data privacy, and optimize the system's energy efficiency, stability, and service life. The specific implementation steps are as follows:

[0102] Step 1: Data collection and preprocessing, at each geographical location (such as Figure 2 The data centers at geographic location 1, geographic location 2, geographic location 3 to geographic location n) regularly collect sensor data of the refrigeration room sensing system, and perform normalization, outlier processing, and quality control on the data;

[0103] Step 2: Local model training. In each data center at each geographic location, the data collected locally is used to train the corresponding health assessment sub-model 22. The key indicators of the health assessment sub-model 22 training include sensor accuracy, data transmission stability, data integrity, equipment failure rate and maintenance frequency.

[0104] Step 3: secure parameter aggregation. Each geographic location transmits the model parameters of the locally trained health assessment sub-model 22 to the central server 11 through encryption, and uses secure multi-party computing technology to aggregate the parameters to ensure the security of the data during the aggregation process.

[0105] Step 4: Global health assessment: the central server 11 performs an overall health assessment of the refrigeration room sensing system based on the aggregated global health assessment model, and provides optimization suggestions and adjustment strategies.

[0106] Step 5: Dynamic adjustment and optimization. Based on the global health evaluation results, the central server 11 sends corresponding optimization suggestions to each geographic location, such as adjusting the operating mode, optimizing the equipment maintenance plan, etc.

[0107] It should be emphasized that the perception system evaluation and optimization method, system, medium, product and terminal based on federated learning provided by the present invention have the following beneficial effects:

[0108] (1) The present invention comprehensively considers multiple key indicators such as sensor accuracy, data transmission stability, data integrity, equipment failure rate, and maintenance frequency to ensure a comprehensive evaluation of the refrigeration room sensing system.

[0109] (2) The present invention makes the evaluation result more accurate through normalization processing and reasonable weight coefficient determination, which helps to more accurately reflect the actual operating status of the perception system.

[0110] (3) The present invention can timely discover potential problems in the system based on the evaluation results, and provide feedback in the form of early warning information to guide maintenance and optimization strategies, thereby improving the reliability and operating efficiency of the system.

[0111] (4) The present invention adopts the federated learning method to achieve practical application in the evaluation and optimization of perception systems across geographical locations, effectively improving the overall performance and reliability of the perception system while protecting data privacy.

[0112] In the embodiments of the present application, words such as "first" and "second" are used to distinguish the same or similar items with substantially the same functions and effects, and do not limit their order. Those skilled in the art can understand that words such as "first" and "second" do not limit the quantity and execution order, and words such as "first" and "second" do not necessarily limit them to be different.

[0113] It should be noted that in the embodiments of the present application, words such as "exemplary" or "for example" represent examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0114] In the embodiments of the present application, "at least one" refers to one or more, and "plurality" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can represent: a, b, c, ab, ac, bc or abc, where a, b, c can be single or multiple.

[0115] Figure 3It is a schematic block diagram of a perception system evaluation and optimization system 300 based on federated learning provided in an embodiment of the present application.

[0116] like Figure 3 As shown, the perception system evaluation and optimization system 300 based on federated learning includes:

[0117] The operation data acquisition module 310 is used to collect the operation data of each geographical location of the refrigeration room sensing system in real time and pre-process the operation data;

[0118] The sub-model construction module 320 is used to construct and train the pre-processed operation data of each geographical location to obtain the health assessment sub-model corresponding to each geographical location;

[0119] A global model acquisition module 330 is used to perform parameter aggregation based on the health assessment sub-models corresponding to each geographical location using the secure multi-party computing technology in the federated learning method to obtain a global health assessment model;

[0120] A health calculation and evaluation module 340 is used to perform calculations based on the health evaluation global model to evaluate the health of the refrigeration room sensing system;

[0121] The optimization and adjustment module 350 is used to optimize and adjust the refrigeration room sensing system according to the evaluation result of the health of the refrigeration room sensing system.

[0122] It should be understood that the specific process of each module executing the above corresponding steps has been described in detail in the above method embodiment, and for the sake of brevity, it will not be repeated here.

[0123] It should also be understood that the division of modules in the embodiments of the present application is schematic and is only a logical function division. There may be other division methods in actual implementation. In addition, each functional module in each embodiment of the present application may be integrated into a processor, or may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated modules may be implemented in the form of hardware or in the form of software functional modules.

[0124] Figure 4 is a schematic block diagram of an electronic terminal provided in an embodiment of the present application. Figure 4 As shown, the computer device includes: at least one processor 401, a memory 402, at least one network interface 403 and a user interface 405. The various components in the device are coupled together through a bus system 404. It can be understood that the bus system 404 is used to achieve connection and communication between these components. In addition to the data bus, the bus system 404 also includes a power bus, a control bus and a status signal bus. However, for the sake of clarity, Figure 4 In the specification, various buses are labeled as bus systems.

[0125] The user interface 405 may include a display, a keyboard, a mouse, a trackball, a click gun, keys, buttons, a touch pad or a touch screen.

[0126] It is understood that the memory 402 can be a volatile memory or a non-volatile memory, and can also include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), which is used as an external cache. By way of example but not limitation, many forms of RAM are available, such as static random access memory (SRAM), synchronous static random access memory (SSRAM). The memory described in the embodiments of the present invention is intended to include but is not limited to these and any other suitable categories of memory.

[0127] The memory 402 in the embodiment of the present invention is used to store various categories of data to support the operation of the electronic terminal 400. Examples of these data include: any executable program for operating on the electronic terminal 400, such as an operating system 4021 and an application 4022; the operating system 4021 includes various system programs, such as a framework layer, a core library layer, a driver layer, etc., for implementing various basic services and processing hardware-based tasks. The application 4022 may include various applications, such as a media player (Media Player), a browser (Browser), etc., for implementing various application services. The perception system evaluation and optimization method based on federated learning provided in the embodiment of the present invention may be included in the application 4022.

[0128] The method disclosed in the above embodiment of the present invention can be applied to the processor 401, or implemented by the processor 401. The processor 401 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit in the processor 401 or the instruction in the form of software. The above processor 401 may be a general processor, a digital signal processor (DSP, Digital Signal Processor), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The processor 401 can implement or execute the various methods, steps and logic block diagrams disclosed in the embodiment of the present invention. The general processor 401 may be a microprocessor or any conventional processor, etc. In combination with the steps of the accessory optimization method provided in the embodiment of the present invention, it can be directly embodied as a hardware decoding processor to execute, or it can be executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium, which is located in a memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0129] In an exemplary embodiment, the electronic terminal 400 may be implemented by one or more application specific integrated circuits (ASIC), DSP, programmable logic device (PLD), complex programmable logic device (CPLD) to execute the aforementioned method.

[0130] According to the method provided in the embodiments of the present application, the present application also provides a computer program product, which includes: computer program code, when the computer program code is run on a computer, the computer executes the perception system evaluation and optimization method based on federated learning of any one of the embodiments shown.

[0131] According to the method provided in the embodiments of the present application, the present application also provides a computer-readable storage medium, which stores a program code. When the program code runs on a computer, the computer executes the perception system evaluation and optimization method based on federated learning of any one of the embodiments shown.

[0132] The terms "component", "module", "system", etc. used in this specification are used to represent computer-related entities, hardware, firmware, a combination of hardware and software, software, or software in execution. For example, a component can be, but is not limited to, a process running on a processor, a processor, an object, an executable file, an execution thread, a program and / or a computer. By way of illustration, both applications running on a computing device and a computing device can be components. One or more components may reside in a process and / or an execution thread, and a component may be located on a computer and / or distributed between two or more computers. In addition, these components may be executed from various computer-readable media having various data structures stored thereon. Components may, for example, communicate through local and / or remote processes according to signals having one or more data packets (e.g., data from two components interacting with another component between a local system, a distributed system and / or a network, such as the Internet interacting with other systems through signals).

[0133] Those of ordinary skill in the art will appreciate that the various illustrative logical blocks and steps described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0134] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0135] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0136] 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, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0137] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0138] In the above embodiments, the functions of each functional unit can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions (programs). When loading and executing computer program instructions (programs) on a computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. Computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, computer instructions can be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media integrations. Available media may be magnetic media (e.g., floppy disks, hard disks, tapes), optical media (e.g., high-density digital video discs (DVDs), or semiconductor media (e.g., solid state disks (SSDs), etc.).

[0139] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0140] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0141] In summary, the present application provides a perception system evaluation optimization method, system, medium, product and terminal based on federated learning, including: real-time collection of operating data of each geographical location of the refrigeration room perception system, and preprocessing of the operating data; constructing and training based on the preprocessed operating data of each geographical location to obtain a health assessment sub-model corresponding to each geographical location; based on the health assessment sub-model corresponding to each geographical location, parameter aggregation is performed using the secure multi-party computing technology in the federated learning method to obtain a global health assessment model; calculations are performed based on the global health assessment model to evaluate the health of the refrigeration room perception system; and the refrigeration room perception system is optimized and adjusted according to the evaluation results of the health of the refrigeration room perception system.

[0142] The present invention comprehensively considers multiple key indicators such as sensor accuracy, data transmission stability, data integrity, equipment failure rate, and maintenance frequency to ensure a comprehensive evaluation of the refrigeration room perception system; the present invention makes the evaluation results more accurate through normalization processing and reasonable weight coefficient determination, which helps to more accurately reflect the actual operating status of the perception system; the present invention can timely discover potential problems in the system according to the evaluation results, and provide feedback in the form of early warning information to guide maintenance and optimization strategies, thereby improving the reliability and operating efficiency of the system; the present invention adopts a federated learning method to realize practical application in the evaluation and optimization of perception systems across geographical locations, effectively improving the overall performance and reliability of the perception system, while protecting data privacy. Therefore, the present application effectively overcomes the various shortcomings of the prior art and has a high industrial utilization value.

[0143] The above embodiments are merely illustrative of the principles and effects of the present application and are not intended to limit the present application. Anyone familiar with the technology may modify or change the above embodiments without violating the spirit and scope of the present application. Therefore, all equivalent modifications or changes made by a person of ordinary skill in the art without departing from the spirit and technical ideas disclosed in the present application shall still be covered by the claims of the present application.

Claims

1. A perception system evaluation and optimization method based on federated learning, characterized in that: include: Collecting the operation data of each geographical location of the refrigeration room sensing system in real time and preprocessing the operation data; Based on the pre-processed operation data of each geographical location, the health assessment sub-models corresponding to each geographical location are constructed and trained respectively; Based on the health assessment sub-models corresponding to each geographical location, secure multi-party computing technology in the federated learning method is used to aggregate parameters to obtain a global health assessment model. Performing calculation based on the health evaluation global model to evaluate the health of the refrigeration room sensing system; The refrigeration room sensing system is optimized and adjusted according to the evaluation result of the health of the refrigeration room sensing system.

2. The perception system evaluation and optimization method based on federated learning according to claim 1 is characterized in that: The health assessment sub-model is constructed by a number of key health indicators; the key health indicators include sensor accuracy, data transmission stability, data integrity, equipment failure rate and maintenance frequency.

3. The perception system evaluation and optimization method based on federated learning according to claim 1 is characterized in that: After normalizing each of the key health indicators and determining the weight coefficient, a weighted sum calculation is performed; the calculation formula of each health assessment sub-model is: SH=k1×SA+k2×DS+k3×DI+k4×ER+k5×MF; Among them, SA is the normalized sensor accuracy, DS is the normalized data transmission stability, DI is the normalized data integrity, ER is the normalized equipment failure rate, MF is the normalized maintenance frequency, and k1, k2, k3, k4, and k5 are the corresponding weight coefficients respectively.

4. The perception system evaluation and optimization method based on federated learning according to claim 3 is characterized in that: The method further comprises: The calculation formula of the normalized sensor accuracy is: SA=D C / D A ; Where SA is the normalized sensor accuracy, D C is the correct number of data items, D T The number of data actually monitored; The calculation formula for the normalized data transmission stability is: DS=D A / D N ; Among them, DS is the normalized data transmission stability, D T is the number of data actually monitored, D N is the standard number of data transmissions; The calculation formula for the normalized data integrity is: DI=D I / D A ; Among them, DI is the normalized data integrity, D I is the number of complete data items, D A The actual number of monitored data.

5. The perception system evaluation and optimization method based on federated learning according to claim 3 is characterized in that: The method further comprises: The calculation formula of the normalized equipment failure rate is: IS=F A / F N ; Among them, ER is the normalized equipment failure rate, F A is the actual failure rate, F N is the standard failure rate; The calculation formula of the normalized maintenance frequency is: MF=M A / M N ; Among them, MF is the normalized maintenance frequency, M A is the actual number of maintenance times, M N This is the standard maintenance frequency.

6. The perception system evaluation and optimization method based on federated learning according to claim 1, characterized in that: The method further comprises: According to the evaluation result of the health of the refrigeration room sensing system, corresponding optimization adjustment suggestions are sent to each geographical location of the refrigeration room sensing system.

7. A perception system evaluation and optimization system based on federated learning, characterized in that: include: An operation data acquisition module is used to collect the operation data of each geographical location of the refrigeration room sensing system in real time and pre-process the operation data; A sub-model construction module is used to construct and train the pre-processed operation data of each geographical location to obtain a health assessment sub-model corresponding to each geographical location; A global model acquisition module is used to aggregate parameters based on the health assessment sub-models corresponding to each geographical location using the secure multi-party computing technology in the federated learning method to obtain a global health assessment model; A health calculation and evaluation module, used for performing calculations based on the health evaluation global model to evaluate the health of the refrigeration room sensing system; The optimization and adjustment module is used to optimize and adjust the refrigeration room perception system according to the evaluation result of the health of the refrigeration room perception system.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the federated learning-based perception system evaluation and optimization method described in any one of claims 1 to 6 is implemented.

9. A computer program product, characterized in that The computer program product includes computer program code, and when the computer program code is executed on a computer, the computer implements the perception system evaluation and optimization method based on federated learning as described in any one of claims 1 to 6.

10. An electronic terminal comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the perception system evaluation and optimization method based on federated learning as described in any one of claims 1 to 6.

Images