Parameter calibration method, operation and maintenance method and medium of device operation and maintenance model

Abstract

The embodiment of the application provides a device operation and maintenance model parameter calibration method, an operation and maintenance method and a medium, and belongs to the parameter calibration field. The method comprises the following steps: obtaining a device operation and maintenance data set; determining a reference valley bottom local feature parameter based on the device operation and maintenance data set; determining a global contour feature parameter according to the reference valley bottom local feature parameter and the device operation and maintenance data set; performing calibration processing on the device operation and maintenance model according to the global contour feature parameter, the reference valley bottom local feature parameter, a first correction formula and a second correction formula to obtain a candidate valley bottom local feature parameter; and outputting a target parameter set under the condition that a preset convergence condition is met. The nonlinear optimization is converted into an efficient linear solution and a correction formula calculation process, the efficiency and the calculation speed of calibration are significantly improved, the precision of parameter calibration is improved, and the data analysis effect of the device operation and maintenance model using the target parameter set for parameter updating is also greatly improved.

Description

Technical Field

[0001] This application relates to the field of parameter calibration technology, and in particular to a parameter calibration method, maintenance method and medium for an equipment operation and maintenance model. Background Technology

[0002] In many fields, such as industry and property management, data analysis based on equipment operation and maintenance models and historical observation data of corresponding scenarios is used to meet the real-time operation and maintenance needs of equipment in different scenarios. Among them, equipment operation and maintenance models often adopt stochastic volatility heuristic models. These models need to ensure the accuracy of parameters to meet the model prediction accuracy. However, in existing technologies, the parameters of equipment operation and maintenance models are often difficult to calibrate, resulting in poor prediction accuracy. Summary of the Invention

[0003] The main objective of this application is to propose a parameter calibration method, operation and maintenance method, and medium for an equipment operation and maintenance model, aiming to improve the accuracy of parameter calibration for the equipment operation and maintenance model, thereby improving the prediction accuracy of the equipment operation and maintenance model updated using the calibrated parameters.

[0004] To achieve the above objectives, a first aspect of this application proposes a parameter calibration method for an equipment operation and maintenance model, the method comprising: Obtain the equipment operation and maintenance model to be calibrated, which is constructed based on the random volatility heuristic model, and the equipment operation and maintenance dataset corresponding to the equipment operation and maintenance model. The equipment operation and maintenance dataset includes multiple sets of operation and maintenance data. Each set of operation and maintenance data consists of equipment operating condition parameters and equipment evaluation indicators that correspond one-to-one with the equipment operating condition parameters. Select the equipment operating parameters when the equipment evaluation index is optimal from the equipment operation and maintenance dataset to obtain the optimal equipment operation and maintenance data; The optimal operating condition parameters and optimal evaluation indicators of the equipment in the optimal operation and maintenance data are used as the horizontal position parameters and smoothness parameters of the operating condition curve of the equipment operation and maintenance model, respectively, to obtain the reference valley local feature parameters. Based on the reference valley bottom local feature parameters and the equipment operation and maintenance dataset, the equipment operation and maintenance model to be calibrated is linearly solved to obtain global contour feature parameters. Based on the global contour feature parameters, the reference valley local feature parameters, the first correction formula, and the second correction formula, the equipment operation and maintenance model is calibrated to obtain the candidate valley local feature parameters for the current iteration. The first correction formula characterizes the relationship between the horizontal position parameter of the operating condition curve to be solved, the smoothness parameter of the operating condition curve, and the optimal operating condition parameter of the equipment when the horizontal position parameter of the operating condition curve is used as the dependent variable. The second correction formula is an implicit equation for the horizontal position parameter of the operating condition curve and the smoothness parameter of the operating condition curve. If the global contour feature parameters and the candidate valley local feature parameters meet the preset convergence conditions, the global contour feature parameters and the candidate valley local feature parameters are used as the target parameter set of the equipment operation and maintenance model to be calibrated and output.

[0005] In some embodiments, after the step of outputting the global contour feature parameters and the candidate valley local feature parameters as the target parameter set of the equipment operation and maintenance model to be calibrated, when the global contour feature parameters and the candidate valley local feature parameters satisfy a preset convergence condition, the method further includes: If the global contour feature parameters and the candidate valley local feature parameters do not meet the preset convergence conditions, the candidate valley local feature parameters of the current iteration are used as the reference valley local feature parameters, and the process jumps to the step of calibrating the equipment operation and maintenance model according to the global contour feature parameters, the reference valley local feature parameters, the first correction formula and the second correction formula to obtain the candidate valley local feature parameters of the current iteration.

[0006] In some embodiments, the global contour feature parameters include initial baseline parameters of the operating condition curve, slope parameters of the operating condition curve, and skew parameters of the operating condition curve. The step of linearly solving the equipment operation and maintenance model to be calibrated based on the reference valley local feature parameters and the equipment operation and maintenance dataset to obtain the global contour feature parameters includes: The horizontal position parameter and smoothness parameter of the operating condition curve of the equipment operation and maintenance model to be calibrated are obtained as linear functions after being assigned the reference valley local feature parameters. The linear function is solved using the least squares method based on the equipment operation and maintenance dataset to obtain the first, second, and third coefficients of the linear function; Based on the first coefficient, determine the initial baseline parameters of the operating condition curve; Based on the third coefficient, determine the slope parameter of the working condition curve. The skewness parameter of the operating condition curve is determined based on the second and third coefficients.

[0007] In some embodiments, the first correction formula is as follows: ; in, Represented as the optimal operating parameters of the equipment. To indicate the first The slope parameter of the operating condition curve in the next iteration, Indicates the first The skewness parameter of the operating condition curve in the next iteration, For the first The horizontal position parameters of the operating condition curve in the next iteration, For the first The smoothness parameter of the operating condition curve in the next iteration.

[0008] In some embodiments, the second correction formula is as follows: ; in, This is represented as the optimal evaluation index for the equipment. Represented as the first Initial baseline parameters of the operating condition curve in the next iteration, For the first The smoothness parameters of the operating condition curve in the next iteration, To indicate the first The slope parameter of the operating condition curve in the next iteration, Indicates the first The skewness parameter of the operating condition curve in the next iteration, For the first The horizontal position parameters of the operating condition curve in the next iteration.

[0009] In some embodiments, after the step of outputting the global contour feature parameters and the candidate valley local feature parameters as the target parameter set of the equipment operation and maintenance model to be calibrated, when the global contour feature parameters and the candidate valley local feature parameters satisfy a preset convergence condition, the method further includes: If there is a skew parameter of the working condition curve that satisfies the preset boundary conditions in the target parameter set, then the coordinate rotation transformation is performed on each group of the operation and maintenance data in the equipment operation and maintenance dataset to obtain the equipment operation and maintenance dataset after coordinate rotation transformation. The equipment operation and maintenance dataset after coordinate rotation transformation is used as the new equipment operation and maintenance dataset. The step of selecting the equipment operating parameters with the optimal equipment evaluation index from the equipment operation and maintenance dataset to obtain the optimal equipment operation and maintenance data is then performed.

[0010] In some embodiments, to enable the analysis of operation and maintenance data, a second aspect of this application proposes an operation and maintenance method for the operation and maintenance of building supporting equipment, the method comprising: Obtain the target business model and the target operation and maintenance data of each building's supporting equipment. The target business model is obtained by updating the equipment operation and maintenance model to be calibrated using the target parameter set obtained through the parameter calibration method of the above-mentioned equipment operation and maintenance model. Operation and maintenance analysis is performed based on the target business model and the target operation and maintenance data.

[0011] In some embodiments, the target operation and maintenance data includes the cumulative runtime of each building's supporting equipment, historical fault records, historical maintenance data, equipment operating environment data, and load rate data. The cumulative runtime of the equipment is the equipment operating condition parameter of the building's supporting equipment. The step of performing operation and maintenance analysis based on the target business model and the target operation and maintenance data includes: Based on the target business model, the cumulative operating time of the device is predicted, and the first failure rate is output. Based on the first model, the historical fault records are predicted, and a second fault rate is output. The first model is obtained by training a bidirectional long short-term memory network model. Based on the second model, the historical maintenance data is predicted, and a third failure rate is output. The second model is obtained by training a model based on a gated cyclic unit model. Based on the third model, the operating environment data of the device is predicted, and a fourth failure rate is output. The third model is obtained by training a model based on a temporal convolutional network. The load rate data is predicted based on the fourth model, and the fifth failure rate is output. The fourth model is obtained by training a gated recurrent unit model with attention mechanism. Based on the equipment type of the building's supporting equipment, the first failure rate, second failure rate, third failure rate, fourth failure rate, and fifth failure rate are weighted and summed to obtain the target failure rate. Operation and maintenance analysis is performed based on the target failure rate.

[0012] In some embodiments, the step of performing operation and maintenance analysis based on the target failure rate further includes: Based on the target failure rates and the historical failure rates of the building equipment, the equipment failure risk level corresponding to each of the building equipment is determined. Among all the building equipment, those whose equipment failure risk level meets the preset risk level conditions are identified as equipment to be maintained; If there are multiple devices to be maintained, obtain the device importance level, current operation and maintenance resources, and maintenance difficulty of each device to be maintained; Based on the target failure rate, equipment importance level, operation and maintenance resources, and maintenance difficulty, the maintenance priority of each piece of equipment to be maintained is determined. Based on the maintenance priority ranking, equipment operation and maintenance strategies are generated to complete the operation and maintenance analysis.

[0013] To achieve the above objectives, a third aspect of the present application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the methods described in the first and second aspects.

[0014] The proposed equipment operation and maintenance model parameter calibration method, operation and maintenance method, and medium acquire an equipment operation and maintenance model to be calibrated based on a random volatility heuristic model, and an equipment operation and maintenance dataset corresponding to the model. The dataset includes multiple sets of operation and maintenance data, each consisting of equipment operating condition parameters and corresponding equipment evaluation indicators. The optimal equipment operating condition parameters are then selected from the dataset to obtain the optimal equipment operation and maintenance data. These optimal parameters and evaluation indicators are used as the horizontal position parameter and smoothness parameter of the operating condition curve of the model, respectively, to obtain reference valley local feature parameters. Finally, based on these reference valley local feature parameters and the equipment operation and maintenance dataset, the parameters are calibrated... The equipment operation and maintenance model to be calibrated is linearly solved to obtain global contour feature parameters. Then, based on the global contour feature parameters, reference valley local feature parameters, a first correction formula, and a second correction formula, the equipment operation and maintenance model is calibrated to obtain candidate valley local feature parameters for the current iteration. The first correction formula represents the relationship between the horizontal position parameter of the operating condition curve to be solved, the smoothness parameter of the operating condition curve, and the optimal operating condition parameter of the equipment when the horizontal position parameter of the operating condition curve is used as the dependent variable. The second correction formula is an implicit equation for the horizontal position parameter of the operating condition curve and the smoothness parameter of the operating condition curve. Finally, if the global contour feature parameters and the candidate valley local feature parameters satisfy the preset convergence conditions, the global contour feature parameters and the candidate valley local feature parameters are used as the target parameter set of the equipment operation and maintenance model to be calibrated and output. It can initialize the local feature parameters of the valley bottom through the optimal operation and maintenance data of the equipment, reducing the dependence on manually setting the initial values. Then, the nonlinear optimization is transformed into an efficient linear solution and correction formula calculation process, which significantly improves the efficiency and calculation speed of calibration. Moreover, the iterative process can converge to the global optimal solution stably and quickly, thereby improving the accuracy of parameter calibration. As a result, the data analysis effect of the equipment operation and maintenance model that uses the target parameter set for parameter update is also greatly improved. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating an embodiment of the parameter calibration method for the equipment operation and maintenance model of this application; Figure 2 This is a flowchart illustrating yet another embodiment of the parameter calibration method for the equipment operation and maintenance model of this application; Figure 3 This is a flowchart illustrating one embodiment of the operation and maintenance method of this application; Figure 4 This is a flowchart illustrating yet another embodiment of the operation and maintenance method of this application; Figure 5 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0017] The steps shown or described are executed sequentially in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0019] The parameter calibration method, electronic device, and storage medium of the equipment operation and maintenance model provided in this application embodiment are specifically described through the following embodiments. First, the parameter calibration method of the equipment operation and maintenance model in this application embodiment is described.

[0020] It should be noted that current mainstream parameter calibration methods for equipment operation and maintenance models using stochastic volatility heuristic models mostly employ traditional nonlinear optimization techniques such as global optimization algorithms and heuristic search. However, these calibration methods have significant technical shortcomings and application limitations: First, they are highly dependent on initial values. When parameter initialization values ​​are unreasonable, the optimization iteration process is prone to getting trapped in local optima, failing to converge to obtain the globally optimal parameter combination, ultimately resulting in large model fitting deviations and insufficient analytical accuracy. Second, they suffer from high computational complexity and low calibration efficiency. Due to the inherent complexity of nonlinear solutions, traditional schemes involve numerous iterations, large computational loads, and slow convergence speeds, making them unsuitable for real-time business analysis needs in industrial and property management scenarios. Third, they exhibit poor overall calibration robustness. Faced with complex and ever-changing business fluctuation scenarios, their parameter generalization ability is weak, further limiting the practical application effectiveness of equipment operation and maintenance models. In summary, existing nonlinear model parameter calibration schemes suffer from a series of problems, including sensitivity to initial values, susceptibility to local optima, low calibration efficiency, insufficient fitting accuracy, and weak robustness, failing to balance calibration accuracy and computational efficiency. Therefore, there is an urgent need to propose a novel method for calibrating equipment operation and maintenance model parameters to reduce dependence on initial parameter values, accelerate convergence speed, improve parameter calibration accuracy and overall robustness, and ensure that the equipment operation and maintenance model can stably and accurately serve various business data analysis scenarios.

[0021] The parameter calibration method for the equipment operation and maintenance model provided in this application relates to the field of parameter calibration technology. The parameter calibration method for the equipment operation and maintenance model provided in this application can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application implementing the parameter calibration method for the equipment operation and maintenance model, but is not limited to the above forms.

[0022] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0023] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirection to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data required for the proper functioning of these embodiments acquired.

[0024] Please see Figure 1 As shown, Figure 1 A flowchart illustrating the parameter calibration method for the equipment operation and maintenance model provided by the present invention includes the following steps: Step S101: Obtain the equipment operation and maintenance model to be calibrated, which is constructed based on the random volatility heuristic model, and the equipment operation and maintenance dataset corresponding to the equipment operation and maintenance model. The equipment operation and maintenance dataset includes multiple sets of operation and maintenance data. Each set of operation and maintenance data consists of equipment operating condition parameters and equipment evaluation indicators that correspond one-to-one with the equipment operating condition parameters. Among them, the equipment operation and maintenance model is a stochastic volatility-inspired model (SVI model). The SVI model is a classic parametric model of the implied volatility surface of options. It can accurately fit the volatility smile shape with only 5 core parameters, satisfy the no-arbitrage constraint, has high calibration efficiency, and is widely used in option pricing, market making, and volatility surface interpolation scenarios. The original expression of the SVI model is as follows: ,in, These are the five core parameters in the SVI model, and the original expression of the equipment operation and maintenance model is the same as that of the SVI model.

[0025] Equipment operating parameters can include operating current, winding temperature, vibration amplitude, load rate, and operating voltage. The corresponding equipment evaluation indicators are current overload rate, temperature exceedance rate, vibration exceedance rate, load overload ratio, and voltage deviation rate. For example, a set of maintenance data could be the equipment's cumulative runtime and the corresponding real-time equipment failure rate. Another set of maintenance data could be the equipment maintenance complexity score and the processing time corresponding to that score.

[0026] In this embodiment, a device operation and maintenance dataset can be read from a specified database or file. The device operation and maintenance dataset may include multiple sets of observation data pairs. ,in, Let be the independent variable, representing the th . The group's equipment operating parameters; Let be the observed value of the dependent variable, representing the th . Equipment evaluation indicators for the group.

[0027] Step S102: Select the equipment operating parameters when the equipment evaluation index is optimal from the equipment operation and maintenance data set to obtain the optimal equipment operation and maintenance data; In this embodiment, curve fitting can be performed on the equipment operation and maintenance dataset to obtain a fitted curve; the minimum point on the fitted curve is taken as the optimal equipment operation and maintenance data, thereby selecting the equipment operating parameters when the equipment evaluation index is optimal. It can be understood that the optimal equipment operation and maintenance data includes the optimal equipment operating parameters and the optimal equipment evaluation index.

[0028] In one embodiment, a quadratic polynomial is used. The coefficients of the fitted curve are obtained by performing least squares fitting on the equipment operation and maintenance dataset. Then, by taking the derivative, the minimum point on the quadratic fitted curve is determined. This minimum point represents the optimal operation and maintenance data for the equipment. For example, for a quadratic fitted curve, its minimum point can be obtained through the derivative: if ,but , The minimum point This data serves as the optimal maintenance data for the equipment under ideal conditions, fitting the curve of the equipment operation and maintenance model, and provides a basis for subsequent parameter initialization. This method automatically determines the baseline point by utilizing the statistical characteristics of the data itself, completely avoiding manual intervention.

[0029] Step S103: The optimal operating condition parameters and optimal evaluation indicators of the equipment in the optimal operation and maintenance data are used as the horizontal position parameters and smoothness parameters of the operating condition curve of the equipment operation and maintenance model, respectively, to obtain the reference valley local feature parameters. The parameters in the equipment operation and maintenance model include valley-level local feature parameters and global contour feature parameters. The valley-level local feature parameters include the horizontal position parameter of the operating condition curve and the smoothness parameter of the operating condition curve. The horizontal position parameter of the operating condition curve is the abscissa corresponding to the minimum point of the equipment's operating condition curve, i.e., the optimal operating condition with the lowest risk. The smoothness parameter of the operating condition curve is the degree of smooth transition of the curve with operating condition fluctuations near the optimal operating condition, determining the curvature of the curve's valley. For example, the horizontal position parameter of the operating condition curve could be the equipment's maximum operating time or maximum load point, and the smoothness parameter of the operating condition curve could be the minimum failure rate of the equipment in its optimal state.

[0030] In this embodiment, the optimal operating condition parameters and optimal evaluation indicators of the equipment in the optimal operation and maintenance data are assigned to the horizontal position parameters and smoothness parameters of the operating condition curve of the equipment operation and maintenance model, respectively, thereby initializing the local feature parameters of the valley bottom, and using the initialized local feature parameters of the valley bottom as reference local feature parameters of the valley bottom.

[0031] Specifically, the optimal operation and maintenance data of the equipment The values ​​are respectively assigned to the horizontal position parameter and the smoothness parameter of the operating condition curve in the local feature parameters of the valley bottom. For example, for the equipment operation and maintenance model, the values ​​can be... , As initial value This represents the horizontal position parameter of the operating condition curve in the 0th iteration. This represents the smoothness parameter of the operating condition curve in the 0th iteration.

[0032] Step S104: Based on the reference valley bottom local feature parameters and the equipment operation and maintenance dataset, perform a linear solution on the equipment operation and maintenance model to be calibrated to obtain global contour feature parameters; In one embodiment, the global profile feature parameters include an initial baseline parameter for the operating condition curve, a slope parameter for the operating condition curve, and a skew parameter for the operating condition curve. The initial baseline parameter for the operating condition curve is used to determine the reference vertical position of the entire operating condition curve. It can be the inherent baseline level of the equipment in its initial factory state; for example, the initial baseline parameter for the operating condition curve can be the basic failure probability at the time of equipment leaving the factory. The slope parameter for the operating condition curve refers to the steepness of the curve as the operating condition changes. It can be the rate at which the operating condition curve rises with the operating time or load increase; for example, the slope parameter for the operating condition curve can be the rate at which the failure rate increases during equipment use. The skew parameter for the operating condition curve is used to control the degree of asymmetric skewness of the curve. The skew parameter for the operating condition curve can be the strength of the correlation between the equipment operating load and the dependent variable of the operating condition curve; for example, the skew parameter for the operating condition curve can be the degree of correlation between the equipment operating load and the failure risk. The value of the correlation can be from 0 to 1, with a correlation of 0 indicating no correlation.

[0033] In one embodiment, step S104 includes: Step S1041: Obtain the horizontal position parameter and smoothness parameter of the operating condition curve of the equipment operation and maintenance model to be calibrated, and assign them as a linear function after referencing the local feature parameters of the valley bottom; In this embodiment, under the condition that the horizontal position parameter and smoothness parameter of the operating condition curve of the equipment operation and maintenance model are assigned as reference valley local feature parameters, the equipment operation and maintenance model is rewritten into a linear form with respect to the global contour feature parameters by using algebraic transformation, thus obtaining a linear function.

[0034] For example, the original form of the equipment operation and maintenance model With fixed reference valley bottom local feature parameters After assigning values ​​to the horizontal position parameter and the smoothness parameter of the working condition curve, it can be transformed using algebraic transformation as follows: In the form of, For linearization parameter set, Given a function, include , , ,in, This is represented as the initial baseline parameter of the operating condition curve. This is expressed as the slope parameter of the operating condition curve. Represented as the skewness parameter of the operating condition curve This is represented as the horizontal position parameter of the operating condition curve. This is expressed as a smoothness parameter for the operating condition curve. Indicates input data, This indicates the output data.

[0035] Step S1042: Solve the linear function using the least squares method based on the equipment operation and maintenance dataset to obtain the first, second, and third coefficients of the linear function; In this embodiment, the linear function can be solved using the least squares method through the equipment operation and maintenance dataset. For example, using the equipment operation and maintenance dataset... Using the least squares method as input and output, this linear form is solved. Least squares is a mature linear regression method that efficiently finds the optimal linear parameters under given conditions. Specifically, it involves constructing the design matrix. and observation vector ,matrix No. The row is represented as Observation vector Based on the actual observations of each historical sample Arranged in sequence, and constructed by solving the normal equations. Obtain the linearized parameter set ,in, Represented as the first The first coefficient, Represented as the first The second coefficient, Represented as the first The third coefficient.

[0036] Step S1043: Determine the initial baseline parameters of the operating condition curve based on the first coefficient; Step S1044: Determine the slope parameter of the working condition curve based on the third coefficient; Step S1045: Determine the skew parameter of the working condition curve based on the second coefficient and the third coefficient.

[0037] In this embodiment, the linearization parameter set It satisfies the relationship with global contour feature parameters For the linearized parameter set The specific values ​​of the initial baseline parameter, the rise slope parameter, and the skew parameter of the working condition curve in the global contour feature parameters are obtained by decomposition. , , .

[0038] Step S105: Based on the global contour feature parameters, the reference valley bottom local feature parameters, the first correction formula, and the second correction formula, the equipment operation and maintenance model is calibrated to obtain the candidate valley bottom local feature parameters for the current iteration. The first correction formula characterizes the relationship between the horizontal position parameter of the operating condition curve to be solved, the smoothness parameter of the operating condition curve, and the optimal operating condition parameter of the equipment when the horizontal position parameter of the operating condition curve is used as the dependent variable. The second correction formula is an implicit equation for the horizontal position parameter of the operating condition curve and the smoothness parameter of the operating condition curve. In this embodiment, the global contour feature parameters, the reference valley bottom local feature parameters, and the optimal operation and maintenance data of the equipment are used to solve the problem using the first correction formula and the second correction formula. The obtained valley bottom local feature parameters are used as candidate valley bottom local feature parameters for the current iteration.

[0039] In some embodiments, the first correction formula is as follows: ; in, Represented as the optimal operating parameters of the equipment. To indicate the first The slope parameter of the operating condition curve in the next iteration, Indicates the first The skewness parameter of the operating condition curve in the next iteration, For the first The horizontal position parameters of the operating condition curve in the next iteration, For the first The smoothness parameter of the operating condition curve in the next iteration.

[0040] In some embodiments, the second correction formula is as follows: ; in, This is represented as the optimal evaluation index for the equipment. Represented as the first Initial baseline parameters of the operating condition curve in the next iteration, For the first The smoothness parameters of the operating condition curve in the next iteration, To indicate the first The slope parameter of the operating condition curve in the next iteration, Indicates the first The skewness parameter of the operating condition curve in the next iteration, For the first The horizontal position parameters of the operating condition curve in the next iteration.

[0041] In one embodiment, the first and second correction formulas can be solved quickly using numerical methods (such as Newton's iteration method) to obtain the local feature parameters of the candidate valley.

[0042] Step S106: If the global contour feature parameters and the candidate valley bottom local feature parameters meet the preset convergence conditions, the global contour feature parameters and the candidate valley bottom local feature parameters are used as the target parameter set of the equipment operation and maintenance model to be calibrated and output.

[0043] In this embodiment, the preset convergence condition can be that the deviation norm (e.g., Euclidean norm) between the candidate target parameters obtained in two adjacent iterations is less than a preset convergence threshold. The candidate target parameters include global contour feature parameters and candidate valley bottom local feature parameters; the preset convergence condition can also be that the current iteration step reaches the preset maximum iteration step.

[0044] In this embodiment, if the global contour feature parameters and the candidate valley local feature parameters meet the preset convergence conditions, it means that the candidate target parameters have been optimized and calibrated. Then, the candidate target parameters are used as the target parameter set of the equipment operation and maintenance model to be calibrated and output.

[0045] In one embodiment, if the global contour feature parameters and the candidate valley local feature parameters do not meet the preset convergence conditions, the candidate valley local feature parameters of the current iteration are used as reference valley local feature parameters, and the process jumps to the step of calibrating the equipment operation and maintenance model according to the global contour feature parameters, the reference valley local feature parameters, the first correction formula and the second correction formula to obtain the candidate valley local feature parameters of the current iteration.

[0046] As an example, if the preset convergence condition is not met, the local feature parameters of the candidate valley in the current iteration are used as the reference local feature parameters of the valley. That is, the local feature parameters of the candidate valley in the current iteration are used as the local feature parameters of the candidate valley in the next iteration. For example, the parameters of the current iteration are used as the reference local feature parameters of the valley. Updated to candidate valley local feature parameters for the next iteration .

[0047] Specifically, let And prepare to enter the next cycle, especially, , , It can be a known parameter that remains unchanged during the iteration process.

[0048] The parameter calibration method for the equipment operation and maintenance model proposed in this embodiment obtains the equipment operation and maintenance model to be calibrated, constructed based on a random volatility heuristic model, and the corresponding equipment operation and maintenance dataset. The equipment operation and maintenance dataset includes multiple sets of operation and maintenance data, each set consisting of equipment operating condition parameters and equipment evaluation indicators corresponding to those parameters. Then, the equipment operating condition parameters at which the equipment evaluation indicators are optimal are selected from the equipment operation and maintenance dataset to obtain the optimal equipment operation and maintenance data. The optimal equipment operating condition parameters and the optimal equipment evaluation indicators from this optimal data are used as the horizontal position parameter and smoothness parameter of the operating condition curve of the equipment operation and maintenance model, respectively, to obtain reference valley local feature parameters. Finally, based on the reference valley local feature parameters and the equipment operation and maintenance dataset, the parameter calibration method is applied to the equipment operation and maintenance model to be calibrated. The equipment operation and maintenance model is solved linearly to obtain global contour feature parameters. Then, based on the global contour feature parameters, reference valley local feature parameters, a first correction formula, and a second correction formula, the equipment operation and maintenance model is calibrated to obtain candidate valley local feature parameters for the current iteration. The first correction formula represents the relationship between the horizontal position parameter of the operating condition curve to be solved, the smoothness parameter of the operating condition curve, and the optimal operating condition parameter of the equipment when the horizontal position parameter of the operating condition curve is used as the dependent variable. The second correction formula is an implicit equation for the horizontal position parameter of the operating condition curve and the smoothness parameter of the operating condition curve. Finally, if the global contour feature parameters and the candidate valley local feature parameters satisfy the preset convergence conditions, the global contour feature parameters and the candidate valley local feature parameters are used as the target parameter set of the equipment operation and maintenance model to be calibrated and output. It can initialize the local feature parameters of the valley bottom through the optimal operation and maintenance data of the equipment, reducing the dependence on manually setting the initial values. Then, the nonlinear optimization is transformed into an efficient linear solution and correction formula calculation process, which significantly improves the efficiency and calculation speed of calibration. Moreover, the iterative process can converge to the global optimal solution stably and quickly, thereby improving the accuracy of parameter calibration. As a result, the data analysis effect of the equipment operation and maintenance model that uses the target parameter set for parameter update is also greatly improved.

[0049] In one embodiment, please refer to Figure 2 As shown, Figure 2 This is a flowchart illustrating a parameter calibration method for an equipment operation and maintenance model according to the present invention. After the step of outputting the global contour feature parameters and the candidate valley local feature parameters as the target parameter set of the equipment operation and maintenance model to be calibrated, provided that the global contour feature parameters and the candidate valley local feature parameters satisfy a preset convergence condition, the method further includes: Step S201: If there is a skew parameter of the working condition curve that satisfies the preset boundary conditions in the target parameter set, then perform coordinate rotation transformation on each group of the operation and maintenance data in the equipment operation and maintenance dataset to obtain the equipment operation and maintenance dataset after coordinate rotation transformation. Step S202: Use the equipment operation and maintenance dataset after coordinate rotation transformation as the new equipment operation and maintenance dataset, and return to the step of selecting the equipment operating condition parameters with the best equipment evaluation index from the equipment operation and maintenance dataset to obtain the optimal equipment operation and maintenance data.

[0050] In this embodiment, it is determined whether there is a skew parameter of the operating condition curve that satisfies the preset boundary conditions in the target parameter set. For example, when the square value of the skew parameter of the operating condition curve is equal to 1, there is a skew parameter of the operating condition curve that satisfies the preset boundary conditions. If there is a skew parameter of the operating condition curve that satisfies the preset boundary conditions in the target parameter set, then each group of operation and maintenance data in the equipment operation and maintenance dataset is subjected to coordinate rotation transformation to obtain a coordinate rotation transformed equipment operation and maintenance dataset. The coordinate rotation transformed equipment operation and maintenance dataset is used as a new equipment operation and maintenance dataset, and the process returns to step S102.

[0051] For example, determining whether there are skew parameters in the target parameter set. satisfy If it exists, then perform a coordinate rotation transformation on the equipment operation and maintenance dataset. Specifically, for each set of operation and maintenance data in the equipment operation and maintenance dataset... A small rotational perturbation is performed to obtain the equipment operation and maintenance dataset after coordinate rotation transformation. This dataset is then used as the new equipment operation and maintenance dataset. The process then returns to the step of selecting the equipment operating parameters with the optimal equipment evaluation indicators from the dataset to obtain the optimal equipment operation and maintenance data. This transforms the boundary problem into a regular problem for solution. It effectively handles parameters at mathematical boundaries (such as...). To address special cases, enhance the robustness and applicability of the algorithm.

[0052] The parameter calibration method for the equipment operation and maintenance model proposed in this embodiment involves performing coordinate rotation transformation on each set of operation and maintenance data in the equipment operation and maintenance dataset when there are skew parameters of the operating condition curve that meet preset boundary conditions in the target parameter set. This results in a coordinate-rotated equipment operation and maintenance dataset. The coordinate-rotated dataset is then used as a new equipment operation and maintenance dataset, and the step of selecting the optimal equipment operating condition parameters from the dataset to obtain the optimal equipment operation and maintenance data is returned. This method transforms boundary problems into regular problems for solution, continuously optimizes the target parameter set, and then uses the target parameter set to update the parameters of the equipment operation and maintenance model to obtain the target business model. Due to the improved accuracy of the target parameter set, the data analysis effect of the target business model is greatly improved.

[0053] In one embodiment, please refer to Figure 3 As shown, Figure 3 This is a flowchart illustrating the operation and maintenance method of the present invention. The operation and maintenance method is used for the operation and maintenance of building supporting equipment, and includes: Step S301: Obtain the target business model and the target operation and maintenance data of each building's supporting equipment. The target business model is obtained by updating the equipment operation and maintenance model to be calibrated with the target parameter set obtained by the parameter calibration method of the equipment operation and maintenance model. Step S302: Perform operation and maintenance analysis based on the target business model and the target operation and maintenance data.

[0054] Among them, the target business model can be a prediction model of equipment failure rate, prediction model of equipment remaining life, prediction model of work order processing time, prediction model of owner satisfaction, or prediction model of park energy consumption in the context of property system; building supporting equipment can be equipment such as underground garage sewage pumps, residential passenger elevators, and community gates.

[0055] In one embodiment, when the target business model is a device failure rate prediction model, each set of operation and maintenance data in the device operation and maintenance dataset includes the device's historical runtime and the failure rate corresponding to the historical runtime. The target operation and maintenance data includes the current cumulative runtime of the device. The device failure rate prediction model is used to output the device failure rate based on the cumulative runtime of the device.

[0056] In one embodiment, when the target business model is a device remaining useful life prediction model, each set of operation and maintenance data in the device operation and maintenance dataset includes the device's historical runtime and the device's remaining useful life corresponding to the historical runtime. The target operation and maintenance data includes the device's current cumulative runtime. The device remaining useful life prediction model is used to output the device's remaining useful life based on the device's current cumulative runtime.

[0057] In one embodiment, when the target business model is a work order processing time prediction model, each set of operation and maintenance data in the device operation and maintenance dataset includes historical work order types and corresponding processing times for historical work order types. The target operation and maintenance data includes the current work order type. The work order processing time prediction model is used to output the work order processing time based on the current work order type.

[0058] In one embodiment, when the target business model is a homeowner satisfaction prediction model, each set of maintenance data in the equipment maintenance dataset includes homeowner historical service records and corresponding satisfaction scores for those records. The target maintenance data includes homeowner current service records. The homeowner satisfaction prediction model is used to output a homeowner satisfaction score based on the homeowner current service records.

[0059] In one embodiment, when the target business model is a park energy consumption prediction model, each set of operation and maintenance data in the equipment operation and maintenance dataset includes the park's historical time period operating conditions and the corresponding energy consumption values ​​for the park's historical time period operating conditions. The target operation and maintenance data includes the park's current time period operating conditions. The park energy consumption prediction model is used to output the park's current time period energy consumption value based on the park's current time period operating conditions.

[0060] The parameter calibration method for the equipment operation and maintenance model proposed in this embodiment obtains the target business model and the target operation and maintenance data of each building's supporting equipment. The target business model is the equipment operation and maintenance model to be calibrated, updated with the target parameter set obtained by the parameter calibration method for the equipment operation and maintenance model. Then, operation and maintenance analysis is performed based on the target business model and the target operation and maintenance data. By using the target business model calibrated and optimized with the target parameter set to conduct operation and maintenance analysis, the method effectively solves the prediction bias problem of traditional models, significantly improves the accuracy of equipment failure risk prediction, and provides a reliable quantitative basis for operation and maintenance decisions.

[0061] In one embodiment, the target operation and maintenance data includes the cumulative runtime of each building's supporting equipment, historical fault records, historical maintenance data, equipment operating environment data, and load rate data, wherein the cumulative runtime of the equipment is the equipment operating condition parameter of the building's supporting equipment.

[0062] Among them, the cumulative operating time of the equipment refers to the cumulative time that the equipment has actually operated under load since it was put into use; for example, the total cumulative operating time of an elevator with passengers since its installation; historical fault records refer to the structured records of all fault events throughout the entire life cycle of the equipment; for example, the occurrence time, fault location, and maintenance records of historical faults of water supply and drainage pumps; historical maintenance data refer to the complete process records of all maintenance, repair, and component replacements throughout the entire life cycle of the equipment; for example, the monthly maintenance records and compressor replacement records of central air conditioning; equipment operating environment data refer to the environmental parameters that affect the operation of the equipment in the installation area; for example, the real-time temperature and humidity, dust concentration, and electromagnetic interference intensity of the power distribution room; load rate data refers to the ratio (in percentage form) of the actual operating load of the equipment to the rated load; for example, the ratio of the real-time load of an elevator to its rated load capacity, and the ratio of the real-time flow rate of a water pump to its rated flow rate.

[0063] The steps for performing operation and maintenance analysis based on the target business model and the target operation and maintenance data include: Step A: Based on the target business model, predict the cumulative operating time of the device and output the first failure rate; The target business model is a stochastic volatility heuristic model.

[0064] Step B: Based on the first model, predict the historical fault records and output the second fault rate, wherein the first model is obtained by training a bidirectional long short-term memory network model; Among them, the Bidirectional Long Short-Term Memory Network (BSSN) is a bidirectional improved recurrent neural network of the Long Short-Term Memory (LSTM) network, capable of processing sequential data simultaneously in both the forward and reverse temporal directions. BSSN is used to capture the bidirectional dependencies and causal relationships within sequential data, accurately uncovering temporal patterns in historical fault records and significantly improving the accuracy of fault rate prediction.

[0065] Step C: Based on the second model, predict the historical maintenance data and output the third failure rate. The second model is obtained by training the model based on the gated cyclic unit model. Among them, the Gated Recurrent Unit (GRU) model is a lightweight improved recurrent neural network that achieves more efficient sequence data processing by simplifying the gating structure. The GRU is used to capture the temporal correlation and intervention impact patterns of long-period, low-frequency sequence data, accurately uncovering the influence of historical maintenance data on equipment status and significantly improving the accuracy of failure rate prediction.

[0066] Step D: Based on the third model, predict the operating environment data of the device and output the fourth failure rate. The third model is obtained by training a model based on a temporal convolutional network. Among them, Temporal Convolutional Network (TCN) is a type of convolutional neural network that can stably cover long-term temporal dependencies. TCN can accurately uncover the impact patterns of equipment operating environment data on equipment aging, significantly improving the accuracy of failure rate prediction.

[0067] Step E: Predict the load rate data based on the fourth model and output the fifth failure rate, wherein the fourth model is obtained by training a gated recurrent unit model with attention mechanism; Among them, the Gated Recurrent Unit with Attention Mechanism (Attention-GRU) is an improved recurrent neural network that integrates the attention mechanism. The Gated Recurrent Unit with Attention Mechanism is used to capture anomalous changes and load shock characteristics in highly volatile time-series data, accurately uncovering the coupling patterns between load rate data and equipment failure risk, and significantly improving the accuracy of failure rate prediction.

[0068] Step F: Based on the equipment type of the building's supporting equipment, perform a weighted summation of the first failure rate, the second failure rate, the third failure rate, the fourth failure rate, and the fifth failure rate to obtain the target failure rate; In this embodiment, based on the equipment's cumulative runtime, historical fault records, historical maintenance data, equipment operating environment data, and load rate data, different models are used to predict the failure rate, resulting in a first failure rate, a second failure rate, a third failure rate, a fourth failure rate, and a fifth failure rate. Then, according to the equipment type of the building's supporting equipment and a preset failure rate weight configuration table, the weight configuration corresponding to each building's supporting equipment is queried in the weight configuration table. The preset weight configuration table is flexibly configured based on the equipment type and the environment in which the equipment is located. For example, the weight of failure rate corresponding to load rate and runtime can be increased for elevator equipment; the weight of failure rate corresponding to environmental corrosion and rainy season load can be increased for underground parking garage water supply and drainage pumps; and the weight of failure rate related to aging can be decreased for newly commissioned equipment, while the weight of failure rate related to load impact can be increased.

[0069] Based on the weight configuration of each of the building's supporting equipment, the first failure rate, second failure rate, third failure rate, fourth failure rate, and fifth failure rate are weighted and summed to obtain the target failure rate.

[0070] It should be noted that the target business model, the first model, the second model, the third model, and the fourth model only have the best prediction accuracy in their respective data dimensions. Weighted summation can integrate the unique advantages of each model and avoid the prediction bias of a single model in an unsuitable dimension.

[0071] As an example, firstly, the cumulative running time, historical fault records, historical maintenance data, equipment operating environment data and load rate data of passenger elevator No. 2 in a certain community are obtained. The probability of occurrence of five faults of the elevator is calculated by the corresponding model. The first fault rate is 3.2%, the second fault rate is 4.5%, the third fault rate is 1.8%, the fourth fault rate is 2.2%, and the fifth fault rate is 5.1%.

[0072] The weighting for passenger elevator scenario adaptation is as follows: Since equipment aging is the core cause of passenger elevator failures, it has the highest weight, configured as the first failure rate weight of 0.35; due to the history of multiple repairs or excessively high repair frequency of elevator No. 2, there is a risk of recurrence, configured as the second failure rate weight of 0.15; since elevator No. 2 recently underwent maintenance, the failure risk is low, configured as the third failure rate weight of 0.15; due to the stable environment in the property scenario, the environmental impact has the lowest weight, configured as the fourth failure rate weight of 0.10; due to the large number of residents in the building where elevator No. 2 is located, the elevator load reaches its peak frequently throughout the day, configured as the fifth failure rate weight of 0.25. In summary, the target failure rate = (3.2% × 0.35) + (4.5% × 0.15) + (1.8% × 0.15) + (2.2% × 0.10) + (5.1% × 0.25) = 3.56%.

[0073] Step G: Perform operation and maintenance analysis based on the target failure rate.

[0074] In this embodiment, after obtaining the accurate target failure rate of each building's supporting equipment, an operation and maintenance strategy is determined based on the accurate target failure rate of each building's supporting equipment to complete the operation and maintenance analysis.

[0075] In one embodiment, please refer to Figure 4 As shown, Figure 4 This is a flowchart illustrating the operation and maintenance method of the present invention. The step of performing operation and maintenance analysis on the target operation and maintenance data based on the target failure rate further includes: Step S401: Based on the target failure rates and the historical failure rates of the building equipment, determine the equipment failure risk level corresponding to each of the building equipment. In this embodiment, the equipment failure risk level can be determined by calculating the ratio between the target failure rate and the historical failure rate. For example, Level 1 is a target failure rate ≥ 8%, or double the historical failure rate; Level 2 is 5% ≤ target failure rate < 8%, or the target failure rate is more than 50% higher than the historical failure rate; Level 3 is 2% ≤ target failure rate < 5%; and Level 4 is a target failure rate < 2%.

[0076] Step S402: Among the various building equipment, identify the building equipment whose equipment failure risk level meets the preset risk level conditions, and designate them as equipment to be maintained; In this embodiment, the equipment that needs to be repaired is selected based on the equipment failure risk level. For example, equipment with a failure risk level of 1 to 3 is selected as the equipment to be maintained.

[0077] For example, the garage sewage pump, fire pump, and No. 1 passenger elevator that meet the requirements are selected from various equipment and designated as equipment to be maintained.

[0078] Step S403: If there are multiple devices to be maintained, obtain the device importance level, current operation and maintenance resources, and maintenance difficulty of each device to be maintained; The equipment importance level is determined by the type of equipment to be maintained. For example, fire protection equipment has the highest importance level, passenger elevators have the second highest importance level, and garage sewage pumps have the third highest importance level. Current maintenance resources refer to the maintenance resources such as on-duty maintenance personnel and spare parts inventory. Maintenance difficulty refers to the difficulty of the equipment maintenance operation.

[0079] Step S404: Based on the target failure rate, equipment importance level, operation and maintenance resources, and maintenance difficulty, determine the maintenance priority ranking of each piece of equipment to be maintained; In this embodiment, each piece of equipment to be maintained can be scored according to a preset maintenance scoring standard, considering its target failure rate, equipment importance level, maintenance resources, and maintenance difficulty, to obtain a maintenance score for each piece of equipment. Based on the maintenance scores of each piece of equipment, they are sorted from highest to lowest to obtain a maintenance priority ranking. For example, the maintenance priority scoring standard is as follows: 1 point for a target failure rate ≥7%, 0.8 points for 5%-7%, and 0.5 points for 3%-5%; Equipment importance level: 1 point for fire / flood control equipment, 0.8 points for passenger elevators / main power distribution equipment, and 0.6 points for general supporting equipment; Maintenance difficulty: 1 point for operation time ≤1 hour, 0.7 points for 1-4 hours, and 0.4 points for more than 4 hours; Maintenance resources: 1 point for complete spare parts + on-duty maintenance personnel, 0.7 points for personnel only, and 0.4 points for spare parts or personnel requiring external transfer. In one embodiment, the maintenance priority ranking is: first, fire pump; second, passenger elevator No. 1; and third, underground parking garage sewage pump.

[0080] Step S405: Based on the maintenance priority ranking, generate equipment operation and maintenance strategies to complete the operation and maintenance analysis.

[0081] In this embodiment, the operation time for each piece of equipment to be maintained is determined based on the maintenance priority of each building's supporting equipment, and an equipment operation and maintenance strategy is generated according to the operation time. For example, the equipment with the highest maintenance priority needs to be repaired immediately. For instance, an equipment operation and maintenance strategy is generated for the first day to maintain the fire pump, the second day to carry out preventive maintenance of the passenger elevator, and the third day to maintain the sewage pump, in order to complete the operation and maintenance analysis.

[0082] The parameter calibration method for the equipment operation and maintenance model proposed in this embodiment determines the equipment failure risk level of each building support equipment based on the target failure rate and the historical failure rate of each building support equipment. Then, it identifies the building support equipment whose failure risk level meets the preset risk level conditions as equipment to be maintained. If multiple equipment are to be maintained, it obtains the equipment importance level, current operation and maintenance resources, and maintenance difficulty of each equipment. Based on the target failure rate, equipment importance level, operation and maintenance resources, and maintenance difficulty, it determines the maintenance priority ranking of each equipment. Finally, based on the maintenance priority ranking, it generates an equipment operation and maintenance strategy to complete the operation and maintenance analysis. This method can accurately screen high-risk equipment to be maintained using target failure rates and historical failure rates, and then integrate multiple dimensions such as failure risk, equipment importance, operation and maintenance resources, and maintenance difficulty to determine maintenance priorities and generate corresponding operation and maintenance strategies, thereby improving the efficiency of operation and maintenance execution.

[0083] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the parameter calibration method of the aforementioned device operation and maintenance model. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0084] Please see Figure 5 , Figure 5 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes: The processor 901 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 902 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 902 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 902 and is called by the processor 901 to execute the parameter calibration method of the device operation and maintenance model of the embodiments of this application. The input / output interface 903 is used to implement information input and output; The communication interface 904 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 905 transmits information between various components of the device (e.g., processor 901, memory 902, input / output interface 903, and communication interface 904); The processor 901, memory 902, input / output interface 903, and communication interface 904 are connected to each other within the device via bus 905.

[0085] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the parameter calibration method of the above-described equipment operation and maintenance model.

[0086] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0087] The parameter calibration method and storage medium for the equipment operation and maintenance model provided in this application embodiment obtains an equipment operation and maintenance model to be calibrated based on a random volatility heuristic model, and an equipment operation and maintenance dataset corresponding to the equipment operation and maintenance model. The equipment operation and maintenance dataset includes multiple sets of operation and maintenance data, each set of operation and maintenance data consisting of equipment operating condition parameters and equipment evaluation indicators corresponding to the equipment operating condition parameters. Then, the equipment operating condition parameters with the optimal equipment evaluation indicators are selected from the equipment operation and maintenance dataset to obtain the optimal equipment operation and maintenance data. The optimal equipment operating condition parameters and the optimal equipment evaluation indicators in the optimal equipment operation and maintenance data are used as the horizontal position parameter and the smoothness parameter of the operating condition curve of the equipment operation and maintenance model, respectively, to obtain the reference valley local feature parameters. Then, based on the reference valley local feature parameters and the equipment operation and maintenance dataset, the parameters of the equipment operation and maintenance model are calibrated. The equipment operation and maintenance model to be calibrated is linearly solved to obtain global contour feature parameters. Then, based on the global contour feature parameters, reference valley local feature parameters, a first correction formula, and a second correction formula, the equipment operation and maintenance model is calibrated to obtain candidate valley local feature parameters for the current iteration. The first correction formula represents the relationship between the horizontal position parameter of the operating condition curve to be solved, the smoothness parameter of the operating condition curve, and the optimal operating condition parameter of the equipment when the horizontal position parameter of the operating condition curve is used as the dependent variable. The second correction formula is an implicit equation for the horizontal position parameter of the operating condition curve and the smoothness parameter of the operating condition curve. Finally, if the global contour feature parameters and the candidate valley local feature parameters satisfy the preset convergence conditions, the global contour feature parameters and the candidate valley local feature parameters are used as the target parameter set of the equipment operation and maintenance model to be calibrated and output. It can initialize the local feature parameters of the valley bottom through the optimal operation and maintenance data of the equipment, reducing the dependence on manually setting the initial values. Then, the nonlinear optimization is transformed into an efficient linear solution and correction formula calculation process, which significantly improves the efficiency and calculation speed of calibration. Moreover, the iterative process can converge to the global optimal solution stably and quickly, thereby improving the accuracy of parameter calibration. As a result, the data analysis effect of the equipment operation and maintenance model that uses the target parameter set for parameter update is also greatly improved.

[0088] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0089] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0090] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0091] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0092] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0093] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0094] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0095] If the integrated unit is implemented as 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 this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0096] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A parameter calibration method for an equipment operation and maintenance model, characterized in that, The method includes: Obtain the equipment operation and maintenance model to be calibrated, which is constructed based on the random volatility heuristic model, and the equipment operation and maintenance dataset corresponding to the equipment operation and maintenance model. The equipment operation and maintenance dataset includes multiple sets of operation and maintenance data. Each set of operation and maintenance data consists of equipment operating parameters and equipment evaluation indicators that correspond one-to-one with the equipment operating parameters. The equipment operating parameters include equipment operating current, winding temperature, vibration amplitude, load rate, and operating voltage. Select the equipment operating parameters when the equipment evaluation index is optimal from the equipment operation and maintenance dataset to obtain the optimal equipment operation and maintenance data; The optimal operating condition parameters and optimal evaluation indicators of the equipment in the optimal operation and maintenance data are used as the horizontal position parameters and smoothness parameters of the operating condition curve of the equipment operation and maintenance model, respectively, to obtain the reference valley local feature parameters. Based on the reference valley bottom local feature parameters and the equipment operation and maintenance dataset, the equipment operation and maintenance model to be calibrated is linearly solved to obtain global contour feature parameters. Based on the global contour feature parameters, the reference valley local feature parameters, the first correction formula, and the second correction formula, the equipment operation and maintenance model is calibrated to obtain the candidate valley local feature parameters for the current iteration. The first correction formula characterizes the relationship between the horizontal position parameter of the operating condition curve to be solved, the smoothness parameter of the operating condition curve, and the optimal operating condition parameter of the equipment when the horizontal position parameter of the operating condition curve is used as the dependent variable. The second correction formula is an implicit equation for the horizontal position parameter of the operating condition curve and the smoothness parameter of the operating condition curve. If the global contour feature parameters and the candidate valley bottom local feature parameters satisfy the preset convergence conditions, the global contour feature parameters and the candidate valley bottom local feature parameters are used as the target parameter set of the equipment operation and maintenance model to be calibrated and output. The first correction formula is shown below: ； in, Represented as the optimal operating parameters of the equipment. To indicate the first The slope parameter of the operating condition curve in the next iteration, Indicates the first The skewness parameter of the operating condition curve in the next iteration, For the first The horizontal position parameters of the operating condition curve in the next iteration, For the first Smoothness parameters of the operating condition curve in the next iteration; The second corrected formula is shown below: ； in, This is represented as the optimal evaluation index for the equipment. Represented as the first Initial baseline parameters of the operating condition curve in the next iteration, For the first The smoothness parameters of the operating condition curve in the next iteration, To indicate the first The slope parameter of the operating condition curve in the next iteration, Indicates the first The skewness parameter of the operating condition curve in the next iteration, For the first The horizontal position parameters of the operating condition curve in the next iteration.

2. The parameter calibration method for the equipment operation and maintenance model according to claim 1, characterized in that, After the step of outputting the global contour feature parameters and the candidate valley local feature parameters as the target parameter set of the equipment operation and maintenance model to be calibrated, provided that the global contour feature parameters and the candidate valley local feature parameters satisfy the preset convergence conditions, the method further includes: If the global contour feature parameters and the candidate valley local feature parameters do not meet the preset convergence conditions, the candidate valley local feature parameters of the current iteration are used as the reference valley local feature parameters, and the process jumps to the step of calibrating the equipment operation and maintenance model according to the global contour feature parameters, the reference valley local feature parameters, the first correction formula and the second correction formula to obtain the candidate valley local feature parameters of the current iteration.

3. The parameter calibration method for the equipment operation and maintenance model according to claim 2, characterized in that, The global contour feature parameters include the initial baseline parameters of the operating condition curve, the slope parameters of the operating condition curve, and the skew parameters of the operating condition curve. The global contour feature parameters are obtained by linearly solving the equipment operation and maintenance model to be calibrated based on the reference valley local feature parameters and the equipment operation and maintenance dataset, including: The horizontal position parameter and smoothness parameter of the operating condition curve of the equipment operation and maintenance model to be calibrated are obtained as linear functions after being assigned the reference valley local feature parameters. The linear function is solved using the least squares method based on the equipment operation and maintenance dataset to obtain the first, second, and third coefficients of the linear function; Based on the first coefficient, determine the initial baseline parameters of the operating condition curve; Based on the third coefficient, determine the slope parameter of the working condition curve. The skewness parameter of the operating condition curve is determined based on the second and third coefficients.

4. The parameter calibration method for the equipment operation and maintenance model according to claim 1, characterized in that, After the step of outputting the global contour feature parameters and the candidate valley local feature parameters as the target parameter set of the equipment operation and maintenance model to be calibrated, provided that the global contour feature parameters and the candidate valley local feature parameters satisfy the preset convergence conditions, the method further includes: If there is a skew parameter of the working condition curve that satisfies the preset boundary conditions in the target parameter set, then the coordinate rotation transformation is performed on each group of the operation and maintenance data in the equipment operation and maintenance dataset to obtain the equipment operation and maintenance dataset after coordinate rotation transformation. The equipment operation and maintenance dataset after coordinate rotation transformation is used as the new equipment operation and maintenance dataset. The step of selecting the equipment operating parameters with the optimal equipment evaluation index from the equipment operation and maintenance dataset to obtain the optimal equipment operation and maintenance data is then performed.

5. An operation and maintenance method for the operation and maintenance of building supporting equipment, characterized in that, The method includes: Obtain the target business model and the target operation and maintenance data of each building's supporting equipment. The target business model is obtained by updating the equipment operation and maintenance model to be calibrated using the target parameter set obtained by the parameter calibration method of the equipment operation and maintenance model as described in any one of claims 1 to 4. Operation and maintenance analysis is performed based on the target business model and the target operation and maintenance data.

6. The operation and maintenance method according to claim 5, characterized in that, The target operation and maintenance data includes the cumulative runtime of each building's supporting equipment, historical fault records, historical maintenance data, equipment operating environment data, and load rate data. The cumulative runtime of the equipment is the equipment operating condition parameter of the building's supporting equipment. The steps for performing operation and maintenance analysis based on the target business model and the target operation and maintenance data include: Based on the target business model, the cumulative operating time of the device is predicted, and the first failure rate is output. Based on the first model, the historical fault records are predicted, and a second fault rate is output. The first model is obtained by training a bidirectional long short-term memory network model. Based on the second model, the historical maintenance data is predicted, and a third failure rate is output. The second model is obtained by training a model based on a gated cyclic unit model. Based on the third model, the operating environment data of the device is predicted, and a fourth failure rate is output. The third model is obtained by training a model based on a temporal convolutional network. The load rate data is predicted based on the fourth model, and the fifth failure rate is output. The fourth model is obtained by training a gated recurrent unit model with attention mechanism. Based on the equipment type of the building's supporting equipment, the first failure rate, second failure rate, third failure rate, fourth failure rate, and fifth failure rate are weighted and summed to obtain the target failure rate. Operation and maintenance analysis is performed based on the target failure rate.

7. The operation and maintenance method according to claim 6, characterized in that, The step of performing operation and maintenance analysis based on the target failure rate further includes: Based on the target failure rates and the historical failure rates of the building equipment, the equipment failure risk level corresponding to each of the building equipment is determined. Among all the building equipment, those whose equipment failure risk level meets the preset risk level conditions are identified as equipment to be maintained; If there are multiple devices to be maintained, obtain the device importance level, current operation and maintenance resources, and maintenance difficulty of each device to be maintained; Based on the target failure rate, equipment importance level, operation and maintenance resources, and maintenance difficulty, the maintenance priority of each piece of equipment to be maintained is determined. Based on the maintenance priority ranking, equipment operation and maintenance strategies are generated to complete the operation and maintenance analysis.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the parameter calibration method of the equipment operation and maintenance model according to any one of claims 1 to 4, or implements the operation and maintenance method according to any one of claims 5 to 7.

Images